xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001913695
007 cr mnu|||uuuuu
008 071015s2007 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001949
Shoemaker, Jonathan Adam.
Construct validity of situational judgment tests :
b an examination of the effects of agreeableness, organizational leadership culture, and experience on SJT responses
h [electronic resource] /
by Jonathan Adam Shoemaker.
[Tampa, Fla] :
University of South Florida,
ABSTRACT: Numerous factors are likely to influence response patterns to situational judgment tests, including agreeableness, leadership style, impression management, and job and organizational experience. This research presents background information and research on situational judgment tests and several constructs hypothesized to influence situational judgment test responses. A situational judgment test and manipulations to influence response patterns were developed and piloted with a small sample of management professionals and undergraduate students. Larger samples of management professionals and undergraduate students participated in the experimental research. Participants were asked to imagine that they are applying for a job. Each participant was presented with background information about a fictitious company, describing a company as either highly Participative/Supportive or highly Directive/Achieving in its leadership culture. A third description provided no information about leadership culture to serve as a control. Participants responded to a situational judgment test consisting of some commercially developed items and some new items. Then participants responded to an inventory comprised of items that measure the factors hypothesized to influence response patterns, specifically Agreeableness and Experience. Significant differences in response patterns were determined to be attributable to the Agreeableness and Experience variables, and the Leadership Culture manipulations, as well as the interaction between Experience and the Leadership Culture manipulations. No significant differences were clearly attributable to the Agreeableness by Leadership Culture interaction. The ramifications of these findings are discussed and recommendations for future research are presented.
Dissertation (Ph.D.)--University of South Florida, 2007.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 100 pages.
Advisor: Michael Brannick, Ph.D.
t USF Electronic Theses and Dissertations.
Construct Validity of Situational Judgment Tests: An Examination of the Effects of Agreeableness, Organizational Leadership Culture, and Experience on SJT Responses by Jonathan Adam Shoemaker A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Psychology College of Arts & Sciences University of South Florida Major Professor: Michael Brannick, Ph.D. Tammy Allen, Ph.D. Walter Borman, Ph.D. Marcia Finkelstein, Ph.D. Joseph V andello, Ph.D. Interdepartmental Chair: Michael Reid, Ph.D. Date of Approval April 11, 2007 Keywords: practical intelligence, tacit knowledge, low fidelity simulation, organizational culture, path goal theory, participative leadership, work experience management, construct explication, job knowledge, personality, agreeableness, personnel selection Copyright 2007, Jonathan Adam Shoemaker
Dedication This research and obtaining my Ph.D. has been a longer process than I ever expected. I would like to thank all my friends and colleagues at the University of South Florida, as well as all the professors who made me think, and all the undergraduates that made me feel like I was making a difference. This paper is dedicated to my wonderful wife Amy Rothman for her support and partnership with this project, during our years in Florida, and always. You are my favorite! I hope I write something more interesting for you next time. Finally, thanks to Beth Â“NanaÂ” Rothman for helping me submit this paper and Â“f anningÂ” the PC.
Acknowledgements I would like to thank my committee for their participation and support throughout this project. I must especially thank my major professor, Dr. Michael Brannick, who was patient and informative to the Â“NthÂ” degree even on days when I was only performing at N 1. Some of my best memories of graduate school will be the theoretical discussions we had about how this project might turn out and what it might all mean. I must also take this opportunity to thank my colleagues at Wa tkins Motor Lines, Inc. for their support of and participation in the pilot portion of this project, my colleagues at Verizon Wireless (notably Ashley Gray and Jon Canger) for their support of and participation in the experimental portion of this project, and the students of the University of South Florida for participating in this project at every level.
i TABLE OF CONTENTS LIST OF TABLES ................................ ................................ ................................ ............ iii LIST OF FIGURES ................................ ................................ ................................ ............ iv ABSTRACT ................................ ................................ ................................ ........................ v Situational Judgment Tests ................................ ................................ ................................ 4 What is a Â“Situational Judgment TestÂ”? ................................ ................................ ......... 4 Reliability of Situational Judgment Tests ................................ ................................ ...... 9 Validity of Situational Judgment Tests ................................ ................................ ........ 10 Criterion Related Validity ................................ ................................ ........................ 10 Construct and Content Validity ................................ ................................ ................ 12 Cognitive Ability ................................ ................................ ................................ .. 12 Interpersonal Traits and Skills ................................ ................................ ............. 13 Face Validity ................................ ................................ ................................ ............ 14 Generalizability of Research on SJTs ................................ ................................ .......... 15 Agreeableness ................................ ................................ ................................ .................. 16 What is Agreeableness? ................................ ................................ ............................... 16 Agreeableness as Antecedent to Performance ................................ ............................. 18 Agreeableness and SJTs ................................ ................................ ............................... 18 Leadership ................................ ................................ ................................ ........................ 19 Why Leadership? ................................ ................................ ................................ ......... 19 Path Goal Theory of Leadership ................................ ................................ .................. 20 Path Goal Theory in Resea rch ................................ ................................ ..................... 22 Impression Management and Faking ................................ ................................ ............... 25 What are Impression Management and Faking? ................................ .......................... 25 Experience ................................ ................................ ................................ ........................ 26 What is Experience? ................................ ................................ ................................ ..... 26 Experience as Antecedent to Performance ................................ ................................ ... 28 Experience and SJTs ................................ ................................ ................................ .... 29 Quantifying Experience ................................ ................................ ............................... 29 Research Hypotheses ................................ ................................ ................................ ....... 30 The Effects of Agreeableness on SJT Scores ................................ ............................... 31 The Effects of Organizational Leadership Culture Manipulations on SJT Responses ................................ ................................ ................................ .............. 32
ii The Interaction of Agreeableness and Organizational Leadership Culture Manipulations on SJT Scores ................................ ................................ ................ 32 The Effects of Experience on SJ T Scores ................................ ................................ .... 33 The Interaction of Job Experience and Organizational Leadership Culture Manipulations on SJT Scores ................................ ................................ ................ 33 Method ................................ ................................ ................................ ............................. 34 Participants ................................ ................................ ................................ ................... 34 Procedure ................................ ................................ ................................ ...................... 36 Measures ................................ ................................ ................................ ...................... 38 Situational Judgment Inventory: The SJT 280 ................................ ........................ 38 Secondary Inventory: The SPE 30 ................................ ................................ ........... 41 Results ................................ ................................ ................................ .............................. 43 Preliminary Analysis: Pilot Testing the Manipulations of ................................ ........... 43 Organizational Leadership Culture ................................ ................................ .............. 43 Preliminary Analysis: Pilot Data ................................ ................................ .................. 44 Preliminary Analysis: Creating a New Frequency Based Key Using Experimental Data ................................ ................................ ................................ 45 Experimental Results: Manipulation Check ................................ ................................ 47 Discussion ................................ ................................ ................................ ........................ 55 Limitations of this Research ................................ ................................ ......................... 58 Directions for Future Research ................................ ................................ .................... 60 References ................................ ................................ ................................ ........................ 65 Appendix A: Instructional Manipulation Information ................................ ..................... 76 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions ...................... 78 Appendix C: Secondary Inventory (SPE 30) and Instructions ................................ ........ 87 Appendix D: Manipulation Pilot Inventory ................................ ................................ ..... 90 Appendix E: Situati onal Judgment Inventory Pilot Rating Instructions .......................... 91 Appendix F: Significant SJT Item level Chi Square and Phi Values .............................. 92 Appendi x G: Item Level Frequency Tables for Managers and Students ......................... 93 Appendix H: Student SJT Item Level Chi Square and Phi Values ................................ .. 99 Ap pendix I: Manager SJT Item Level Chi Square and Phi Values ............................... 100 About the Author ................................ ................................ ................................ ............ 101
iii LIST OF TABLES Table 1: Reliability Estimates of Situational Judgment Tests in Various Studies ............. 9 Table 2: Relationships Between SJT Scores and Supervisor Ratings of Performance .... 11 Table 3: Comparative Experience for Managers vs. Students ................................ ......... 36 Table 4: Interpretations of Leadership Characteristics Between Manipulation Co nditions ................................ ................................ ................................ ................. 43 Table 5: Manipulation Check Composite Scores Among 3 Conditions .......................... 47 Table 6: Means, Standard Deviations and Score R anges for Agreeableness ................... 49 Table 7: SJT Score Means and Standard Deviations by Manipulation ............................ 50 Table 8: Means, Standard Deviat ions and Correlations Between Agreeableness & SJT Score by Condition ................................ ................................ ........................... 51 Table 9: Mean SJT Scores and Standard Deviations by Experience Status .................... 52 Table 10: Mean SJT Scores and SDs by Experience and Leadership Condition ............. 53 Table 11: ANCOVA Model of Main Effects and Intera ctions ................................ ........ 55
iv LIST OF FIGURES Figure 1: A Nomological Network of Constructs Affecting SJT Response .................... 30 Figure 2: Agreeableness X Leadership Characteristics Interaction Effects on SJT Scores ................................ ................................ ................................ ........................ 33 Figure 3: Experience X Leadership Characteristics Interaction on SJT Response .......... 34 Figure 4: Slopes of Agreeableness SJT Score Relationships by Leadership Culture ................................ ................................ ................................ ....................... 52 Figure 5: Student & Manager SJT Scores Across Conditions ................................ ......... 54
v Construct Validity of Situational Judgment Tests: An Examination of the Effects of Agreeableness, Organizational Leadership Culture and Experience on SJT Responses Jonathan Ada m Shoemaker ABSTRACT Numerous factors are likely to influence response patterns to situational judgment tests, including agreeableness, leadership style, impression management, and job and organizational experience. This research presents background info rmation and research on situational judgment tests and several constructs hypothesized to influence situational judgment test responses. A situational judgment test and manipulations to influence response patterns were developed and piloted with a small s ample of management professionals and undergraduate students. Larger samples of management professionals and undergraduate students participated in the experimental research. Participants were asked to imagine that they are applying for a job. Each part icipant was presented with background information about a fictitious company, describing a company as either highly Participative/Supportive or highly Directive/Achieving in its leadership culture. A third description provided no information about leaders hip culture to serve as a control. Participants responded to a situational judgment test consisting of some commercially developed items and some new items. Then participants responded to an inventory comprised of items that measure the factors hypothesi zed to influence response patterns,
vi specifically Agreeableness and Experience. Significant differences in response patterns were determined to be attributable to the Agreeableness and Experience variables, and the Leadership Culture manipulations, as well as the interaction between Experience and the Leadership Culture manipulations. No significant differences were clearly attributable to the Agreeableness by Leadership Culture interaction. The ramifications of these findings are discussed and recommenda tions for future research are presented.
1 Construct Validity of Situational Judgment Tests: An Examination of the Effects of Agreeableness, Organizational Leadership Culture and Experience on SJT Responses The ultimate purpose of selection testing is to provide a prediction of performance that is accurate and inexpensive. Accuracy is important because well designed tests should be scientifically developed and validated to fairly exclude unqualified candidates and to select those that will perform best Ideally, tests should be highly accurate in the settings for which they were developed, and they should be easily transferable to new settings. Affordability is desirable because for applied purposes, it is difficult to implement a high priced, cumbers ome test that requires a major commitment of resources. Situational judgment tests (SJTs) may fit both criteria. SJTs are face valid; that is, the stimuli and responses required by these tests appear to be related to activities that are performed on the job. High face validity coupled with good predictive validity and the relatively inexpensive nature of such tests has made SJTs a popular selection technique. However, the construct(s) being measured by situational judgment tests has received very little attention, and our understanding of what the tests actually measure is very shallow. This research strives for better understanding the nature of the constructs that affect responses to situational judgment tests. It was hypothesized that multiple factor s influence responses to situational judgment tests. Among these are: Agreeableness, as
2 described by McCrae and Costa (1985) in their Five Factor Model, The Leadership Culture of an organization, specifically in terms of manipulations based on Path Goal T heory, as proposed by House (1971), motives of the test taker (intent to present a favorable impression), and job experience. There is already great deal of evidence that situational judgment tests are related to cognitive ability, experience, and even pe rsonality. However, it is still unclear what other constructs or variables may provide incremental variance in relation to SJT responses, and how all these constructs work together to affect response choices. This study proposed that a nomological repres entation of responses to situational judgment items, specifically those that address dealing with subordinates, should also include agreeableness on the part of the respondent, impression management on the part of the respondent, the leadership characteris tics of the organization, and job experience, as well as interactions between these constructs. This paper begins by presenting background information and research on situational judgment tests themselves, their reliability and validity, and how they are u sed. Then the paper describes those constructs hypothesized to influence SJT responses and the rationale for their inclusion in the current research. First, Agreeableness (and its opposite, Antagonism) is described. Next, the concept of an organizationa l leadership culture will be discussed, with emphasis on Path Goal Theory (House, 1974). Some attention will be paid to the importance of impression management in the workplace Â– specifically, respondentsÂ’ attempts to display congruence between their own style of
3 leadership and the organizationÂ’s leadership culture. Finally, the concept of experience will be explored. A rudimentary explanation of the study is described below so that the literature review and hypotheses that follow can be placed in context Participants from two settings were included in the research: the professional or Â“ManagerÂ” sample included management employees, while the novice or Â“StudentÂ” sample included college students. Participants were asked to imagine they are applying for a job. Each participant was presented with one of three sets of background information about a fictitious company. One set of information clearly indicated that the company is highly Participative and Supportive in its leadership culture. Another set of information clearly indicated that the company is highly Directive and Achievement Oriented in its leadership culture. The third set of information contained no information about the leadership culture and acted as a control. Participants then responded to a situational judgment test. Finally, participants responded to an inventory that included items that address personality and experience, as well as manipulation check items. Simply put, participants were expected to respond to the situational judgm ent test differently based on the variables mentioned above: notably, agreeableness on the part of the respondent, impression management on the part of the respondent, the leadership characteristics of the organization (as operationalized by the manipulati ons of background information), and experience in the job of manager.
4 Situational Judgment Tests What is a Â“Situational Judgment TestÂ”? A situational judgment test is essentially a work sample test that presents job specific problems to which a test taker must choose from among several possible solutions. The concept of using a work sample as an assessment tool has been around since the beginnings of the field of personnel psychology. Work samples have actually demonstrated the highest validity for any individual assessment method, even higher than that of cognitive ability tests (Schmidt & Hunter, 1998). Tests that present written situations to simulate actual work were first used in selection in the 1920Â’s (McDaniel & Nguyen, 2001). The first inst rument to be classified as a situational judgment test was developed in the late 1950Â’s to help select supervisors (Mowry, 1957). Situational judgment tests are a natural progression from more elaborate simulation techniques such as assessment centers and other work simulations. Because the situational fidelity is compromised when the respondent is not physically in the situation, SJTs have been referred to as Â“low fidelity simulations,Â” a term coined by Motowidlo, Dunnette, and Carter (1990). Situationa l judgment tests have been criticized as Â“a rather extreme attempt to streamline assessment centersÂ” (Borman et al., 1997, p. 314). However, other authors suggest that situational approaches to selection techniques are one of the most significant developme nts in selection research (Robertson & Smith, 1989). Situational judgment tests continue to be popular because they appear to be highly work related, more so than many personality or cognitive ability tests. Therefore, they
5 demonstrate better face val idity and more favorable reactions from respondents (Ryers & Connerly, 1993). SJTs have also been shown to have less adverse impact than tests of cognitive ability ( e.g. Oswald, et al 2004; Callinan & Robertson, 2000; Chan & Schmidt, 1997; Weekley & Jon es, 1997; Motowidlo & Tippins, 1993). SJTs are characterized by a stem/response format (McDaniel & Nguyen, 2001). Each item begins with a stem that presents a work related situation. Then, a series of response options are presented. Stems and responses can vary in fidelity, length and complexity (McDaniel & Nguyen, 2001). Typical SJTs are presented and completed in paper and pencil format, although a few SJTs use a video format in which respondents watch videotaped scenarios and choose from a set of vi deotaped or written responses (Weekley & Jones, 1997). Most situational judgment tests are developed for specific companies or for specific jobs or job families (Hanson & Ramos, 1996). There are few commercially available SJTs. Two notable exceptions ar e Â“The ProveIt! Manager,Â” published by Kenexa, Inc., and the Â“Supervisory Skills InventoryÂ” (SSI), published by gNeil, Inc., though only portions of these two tests include situational judgment items. Items from these tests were included in this research and information about them is elaborated below. SJTs often consist of a smaller number of items than are commonly seen on cognitive ability or personality instruments. Many SJTs consist of as few as 20 item stems (Hanson & Ramos, 1996). This is primarily because every item requires the respondent to read detailed situational stems before responding. Response choices
6 usually occur along a continuum, ranging from behavior(s) considered most effective to those considered least effective. SJTs are most of ten developed from critical incidents (McDaniel & Nguyen, 2001). Next, response options are created when subject matter experts and/or novices unfamiliar with the job generate lists of effective and ineffective reactions to the described situation. Final ly, scoring keys are developed, either rationally, by asking experts to rate the effectiveness of each response, or empirically by having participants take the test and comparing their scores to some external criterion, such as performance (Hanson & Ramos, 1996). A widely accepted technique for situational judgment test instructions is to ask for both an effective and ineffective response, as introduced by Motowidlo, Dunnette and Carter (1990). This method provides information not only on effective perform ance, but also on the ability to avoid those most severely ineffective behaviors (Hanson & Ramos, 1996). Instructions for situational judgment tests may also address a respondentÂ’s potential or actual behavior. For example, one of the most popular scorin g formats, as proposed by Motowidlo, Dunnette and Carter (1990) is for respondents to indicate which response choices are the most effective and which the least effective. Motowidlo and McDaniel (2005) referred to this scoring format as Â“knowledge instruc tions,Â” as contrasted with Â“behavioral tendency instructions,Â” that ask respondents to indicate which response choice is most like them and which is least like them. Other researchers have called these opposing instructional formats Â“should doÂ” and Â“would do,Â” respectively (Ployhart & Ehrhart, 2003). Findings suggest that Â“would doÂ” or behavioral
7 tendency instructions show better reliability and validity than Â“should doÂ” or knowledge instructions (Ployhart & Ehrhart, 2003). However, it is also likely tha t Â“should doÂ” instructions are more appropriate for addressing maximal performance such as what would be expected from job applicants (Ployhart & Ehrhart, 2003). Â“Should doÂ” and knowledge instructions appear to allow less dissimulation, and do not inflate scores when respondents are motivated to fake (McDaniel & Nguyen, 2001), such as they might be when applying for a job. The instructional format used also has implications for how much responses are influenced by cognitive ability vs. non cognitive trait s. SJTs with knowledge instructions tend to be more highly correlated with general cognitive ability; SJTs with behavioral tendency instructions tend to be more highly correlated with non cognitive traits such as Conscientiousness and Agreeableness (McDan iel, Grubb & Hartman, 2003). Situational judgment tests can vary widely in length, instructions and stem/response format. Regardless of the format, SJTs are intended to predict performance; however it remains unclear exactly what construct(s) SJTs truly are measuring. There has been a rift in the literature as some researchers suggest situational judgment is a truly unique construct, also referred to as tacit knowledge (Sternberg et al ., 1995) or procedural knowledge. Other researchers consider situatio nal judgment tests to be merely another method of testing already well established constructs such as job knowledge, work experience, or even general cognitive ability (Stevens & Campion, 1999; Chan & Schmidt, 1997; Schmidt & Hunter, 1993).
8 Sternberg and his associates (1995) suggested that SJTs measure a unique construct that is not related to general cognitive ability, or even job knowledge, as it is traditionally operationalized. Wagner (1987) called this construct tacit knowledge, a practical intellig ence that is never described formally or taught directly within an organization. According to SternbergÂ’s triarchic theory of intelligence, practical intelligence exists outside of the traditional sphere of general cognitive ability (Sternberg, 1985). Ta cit knowledge is typically procedural and goal oriented (Sternberg & Wagner, 1993). Further, tacit knowledge is acquired without formal instruction from others (Sternberg et al. 1995). Sternberg has equated practical or tacit knowledge with Â“street smar ts,Â” Â“learning the ropesÂ” and Â“common senseÂ” (Wagner & Sternberg, 1991, p. 1). One major drawback to tacit knowledge is plain from its very name: unspoken, unofficial information transfer is very difficult to measure. The result is that researchers stil l write about tacit knowledge as a theoretical construct while also describing it as synthetic, intuitive, and not easy to operationalize (Styhre, 2004). Chan and Schmidt (1997) argued that a test of situational judgment is simply a method of measur ing multiple job relevant skills and abilities. Studies of biodata, interviews and assessment centers have demonstrated that these are methods of measuring common constructs, not constructs in and of themselves (Schmidt & Rothstein, 1994). The same is al most certainly true of SJTs. The balance of opinion now suggests that whether tacit knowledge exists or not is immaterial. Current theory considers the situational judgment test to be a style of measurement that taps numerous constructs, not a single, spe cific construct. However,
9 little research has been performed to show how the constructs measured by SJTs are affected by other constructs (Motowidlo & McDaniel, 2005). Reliability of Situational Judgment Tests Several researchers have developed unique sit uational judgment instruments. Most report reliability coefficients in the form of internal consistency. This statistic estimates the correlation that would be observed if the examinees took another test Â‘just like this oneÂ’ and the correlation between t he (often hypothetical) alternate forms was computed. Table 1 shows internal consistency reliabilities (CronbachÂ’s alpha ) for SJTs used in research. A few studies reported additional measures of reliability, in lieu of, or in addition to, internal consis tency. The additional reliability coefficients included in the table are inter rater reliability, test retest reliability, and alternate forms reliability. Table 1: Reliability Estimates of Situational Judgment Tests in Various Studies Study N N of SJT I tems Reliability Coefficient(s) Lee, Choi & Choe (2005) 498 16 = .13 .64* Oswald et al. (2004) 634 57 = .85 Ployhart et al. (2003) 5325 10 = .46 .62* Ployhart & Ehrhart (2003) 84; 23 5 = .36; test retest = .63 Chan & Schmidt (2002) 160 8 = 73; alternate forms = .76 Clevenger et al. (2001) 412 39 = .63 .82* Motowidlo, Dunnette & Carter (1990) 252 58 Mean inter rater Reliability = .95 Weekly & Jones (1999) 1884 34 = .73 Clause et al. (1998) 377 33 Alternate forms = .70 .77* *Ranges o f reliability are shown if more than one sample or form was used in the research As illustrated above, internal consistency reliability coefficients have ranged from poor to acceptable, according to the current standard of acceptable reliability for use in research (.70; Nunnally & Bernstein, 1994). The principles of test theory state that a larger number of items will increase reliability. However, one of the tests with the fewest
10 items and the smallest number of participants still demonstrated acceptabl e internal consistency reliability (Chan & Schmidt, 2002). This may be because the items addressed one specific construct instead of numerous situational judgment constructs. It appears possible to construct a situational judgment test with acceptable i nternal consistency with as few as 15 to 30 items, especially if that test addresses a single construct of interest ( e.g. dealing with subordinates). Validity of Situational Judgment Tests The validity of a test shows the degree to which the test is us eful in light of the inferences that the test giver wishes to make. Simply put, a situational judgment test is our best guess about actual situational judgment, or how respondents might respond in a real life situation. While no test is a completely accu rate reflection of behavior, tests can be shown to be significant predictors of behavior and/or be related to important factors that affect behavioral outcomes. Psychologists refer to test validity as being demonstrated both externally and internally. E xternal validity refers to what the test can meaningfully predict in a practical sense. External validity is chiefly demonstrated through criterion related validity, where a criterion is some real world measure of performance that can be shown to meaningf ully relate to how a respondent performs on a test. Internal validity refers to the nature of the test itself: which construct(s) the test measures, what the content domain of the test is, and how respondents perceive the test. Criterion Related Validit y Most research investigating criterion related validity has used highly subjective criteria, such as multi faceted performance ratings from supervisors. A meta analysis by
11 McDaniel, Morgeson, Finnegan, Campion, and Braverman (2001) showed that SJTs have a mean uncorrected validity of .26 with performance. Numerous studies have shown a moderate correlation between SJT responses and supervisor performance ratings (Chan & Schmidt, 2002; Clevenger et al. 2001; Weekley & Jones, 1997; Motowidlo & Tippins, 199 3; Motowidlo, Dunnette & Carter, 1990) and turnover (DÂ’Alessio, 1994). Table 2 shows several studies that reported correlations between SJT scores and subjectively rated job performance. Table 2: Relationships Between SJT Scores and Supervisor Ratings of Performance Study N Coefficient ( r ) Chan & Schmidt (2002) 160 .30* Clevenger et al. (2001) 412 .21* (avg.) Motowidlo & Tippins (1993) 165 .31* Motowidlo, Dunnette & Carter (1990) 252 .30** Weekly & Jones (1997) 1471 .35** *Significant at p < .05, ** significant at p < .01 Oswald et al. (2004) similarly determined that SJT scores correlated with self ratings of college performance in a sample of college freshmen. Stevens and Campion (1999) developed an SJT that addresses teamwork. They determined tha t their instrument was moderately correlated with ratings of teamwork performance ( r = .32) and overall performance ( r = .37), though their instrument was highly redundant with the measure of general cognitive abilities that they used in the same study ( r = .95). Weekley and Jones (1999) determined that their situational judgment instrument was correlated only weakly ( r = .19, n.s. ) with performance ratings, despite the fact that their previous research showed more predictive validity.
12 Construct and Con tent Validity Construct validity refers to an instrumentÂ’s ability to measure an unobservable idea or construct. Constructs must be defined through the characteristics of the variables that are used to measure them. The most common way to demonstrate the construct validity of an instrument is to compare it to other, well established measures of the construct. Content validity refers to how completely an instrument addresses the entire realm of characteristics that make up a construct. The Â“realm of char acteristicsÂ” is more appropriately called the content domain. A substantial portion of SJT research has suggested that SJT responses are little more than measures of general cognitive ability ( g ). Cognitive Ability Weekly and Jones (1999) showed that SJ Ts are significantly related to cognitive ability with an average weighted correlation of .45, however, their sample of Yale undergraduates and managerial employees certainly had a severe restriction of range on the cognitive ability measure. McDaniel et al. (2001) provided meta analytic evidence that SJT responses have an average corrected correlation of .39 with general cognitive ability. This finding shows the clear import of cognitive ability in understanding SJT responses; it also shows that there is more to SJT responses than just ( g ). For example, McElreath and Vasilopoulos (2002) reported that Â“most likelyÂ” and Â“least likelyÂ” responses to SJT items have a different relationship with ( g ): least likely SJT scores had a stronger relationship with cog nitive ability than did most likely SJT scores. This may be because what should not be done is typically very clear to respondents with higher
13 cognitive abilities, although what should be done is not always as obvious. Further, Motowidlo, Dunnette and C arter showed that SJT scores did not correlate with aptitude test scores (1990). Numerous studies have shown that SJTs provide incremental validity over and above cognitive ability tests (Chan & Schmidt, 2002; Clevenger et al. 2001; McDaniel et al. 2001 ). The incremental validity provided by SJTs affords the notion that there is more to the content domain of the SJT than just general cognitive ability. The content domain of situational judgment response may be incomplete because too much emphasis is p laced on ( g ). While cognitive ability is certainly an important factor in understanding situational judgment response, considering only cognitive ability ignores the possible influence of Agreeableness, multifaceted levels of experience, and other variabl es. Interpersonal Traits and Skills McDaniel and Nguyen (2001) performed a meta analysis in which they determined that Agreeableness correlated with SJT responses (mean r = .25 over 12 studies). SJT scores have also shown to be correlated with Â“interperso nal skills,Â” Â“communication skills,Â” & Â“negotiation skillsÂ” as rated by interviewers (Motowidlo, Dunnette & Carter, 1990). There has also been some exploration of how responses to SJTs may be influenced by personality, not by levels of a given trait specif ically, but by how effective different levels of expression of that trait are perceived to be in a given situation. For example, Motowidlo and colleagues (2006) coined the term ITP or implicit trait policy, suggesting that personal levels of Agreeableness could influence responses, but also that
14 the amount of importance given to Agreeableness in decision making would influence how favorably participants viewed high Agreeableness versus low Agreeableness response options in terms of effectiveness. Motowidl o and colleagues (2006) showed that personality traits do have some influence on situational judgment; notably, procedural knowledge scores were significantly correlated with agreeableness scores r = .25 ( p < .01), and procedural knowledge scores were sig nificantly correlated with implicit trait policy for Agreeableness r = .73 ( p < .01). Borman and his colleagues (1991) reported that job knowledge mediates the relationship between cognitive ability and performance. Other researchers have agreed th at job knowledge (as operationalized by job experience) appears to have a positive relationship with SJTs (McDaniel & Nguyen, 2001). Weekly and Jones (1999) found that SJTs are correlated with overall job experience ( r = .23), but not necessarily with com pany tenure ( r = .02). Face Validity One of the main reasons that Situational judgment tests are highly valued is because they typically show a high degree of face validity. Face validity may be the most important kind of validity in terms of creating an instrument that encourages valid responses from the test taker. Face validity is a theoretical term that refers to how the instrument appears, especially to the respondent. Situational judgment tests offer a high degree of face validity because responden ts perceive the items on the tests to be highly related to the duties they will perform on the job. A test of general cognitive ability may not be perceived in the same way, because the items on a cognitive ability test, though
15 predictive of performance o utcomes, may be perceived as being outside the context of realistic job performance (Callinan & Robertson, 2000). Likewise, the perception that a test is an opportunity to show how well they can perform on highly job related tasks leads to more applicant motivation and better engagement in the test (Callinan & Robertson, 2000). Generalizability of Research on SJTs Situational judgment tests appear most valid for the jobs and organizations for which they were originally developed. Even though similar situ ations may occur in different organizations, differences in organizational goals, culture or values may require unique scoring keys (Hanson & Ramos, 1996). Research on the generalizability of SJTs across organizations would be a great contribution to the literature (Hanson & Ramos, 1996). It is widely accepted that a high level of cognitive ability is desirable for success in every organization. However, it is the basis of this research that SJTs cannot fairly be generalized across organizations because different organizations require different degrees of characteristics such as Agreeableness and leadership style. The body of empirical research offers numerous recommendations for future research using situational judgment tests. Weekly and Jones (1999 ) and McDaniel and colleagues (2001) suggested that a nomological net of the constructs that SJTs measure should be developed, and recommended measuring specific constructs with SJTs by developing items directly related to those constructs. These authors also suggested that SJTs are most likely to have different nomological nets (thus, address different constructs) if they are based on unique aspects of job content in different jobs; for
16 example, dealing with subordinates would probably have a different no mological net than closing the sale with new clients, even though both types of items may appear on the same situational judgment test. McDaniel and Nguyen (2001), Clevenger and his colleagues (2001), Chan and Schmidt (2002), and Motowidlo and his colleag ues (2006) echoed the recommendation that future research should develop strategies to target specific constructs within the context of situational judgment items. Ployhart and Ehrhardt (2003) suggested that more research is needed on the psychological pr ocesses that people employ to complete SJTs. It has also been suggested that experience should be measured more specifically to better understand the role it plays in SJT response (Chan & Schmidt, 2002; Clevenger et al. 2001). Finally, Oswald and collea gues (2004) called for a better understanding of how impression management affects responses to situational judgment items. The current research moves in these directions first by narrowing the focus of an SJT to issues of supervisors dealing with subordi nates and second by elaborating the nomological net. Specific variables to be included in the nomological net are described next. Agreeableness What is Agreeableness? As early as the 1960Â’s, personality researchers had developed a rudimentary Â“Five Fact orÂ” model of human personality (Norman, 1963; cited in McCrae & Costa, 2003). More than 20 years later, McCrae and Costa (1985) first proposed the Â“Big FiveÂ” model that is widely accepted today. The five factors that Costa and McCrae hypothesized are:
17 Ex troversion, Agreeableness, Conscientiousness, Neuroticism, and Openness to Experience. Agreeableness is described as selflessness, concern for others, trust and generosity of sentiment (McCrae & Costa, 2003). The antithesis of Agreeableness is referred to as Antagonism or tough mindedness (McCrae & Costa, 2003). It should be noted that a high degree of Agreeableness is not always desirable; in some cases, less Agreeableness, or even Antagonism is advantageous ( e.g. for prosecuting attorneys, soldiers in combat conditions, or simply committee members who do not wish to take on additional responsibilities). McCrae, Costa, and Busch (1986) described the Agreeable and Antagonistic personality through comparison with a set of personality aspects called the California Q Set. They described a highly Agreeable individual as: sympathetic, considerate, warm, compassionate, arousing, liking and behaving in a giving way. They described a highly Antagonistic individual as: critical, skeptical, showing condescensi on, pushing limits and expressing direct hostility. Both of these types may be advantageous in certain settings. These type descriptions were used to aid in creating instructional manipulations for this research. Costa, McCrae and Dye (1991) specified six facets that fall under the factor of Agreeableness. These six facets are: Trust, Straightforwardness, Altruism, Compliance, Modesty, and Tender mindedness. Each of these has implications for use in the workplace, and were addressed specifically in me asures used for this research.
18 Agreeableness as Antecedent to Performance For almost as long as the Five Factor Model of personality has been established, researchers have explored personality variables as predictors of performance. Tett and colleagues reported in their meta analysis of confirmatory personality and performance studies that the correlation between Agreeableness and performance was the highest of all the Big 5 personality variables (average r = .326; Tett, Jackson & Rothstein, 1991). Thi s is in contrast to a similar meta analysis performed the same year that showed little relationship between Agreeableness and performance, even in jobs that seem to require a high degree of sociability (Barrick & Mount, 1991) 1 Further research later conc eded that Agreeableness (among other personality variables) is a better predictor of job performance in highly autonomous jobs, such as management, than in less autonomous jobs, though the magnitude of the reported correlations remained small (Barrick, Mou nt & Judge, 2001; Barrick & Mount, 1993). Agreeableness and SJTs Although evidence of a significant relationship between Agreeableness and performance is slim, some research has supported the potential relationship between Agreeableness and responses to situational judgment items. McDaniel and Nguyen (2001) showed that SJTs are correlated on average with measures of Agreeableness ( r = .25), Neuroticism ( r = .31), and Conscientiousness ( r = .27). Cucina, Vasilopoulos and Leaman (2003) suggested that Â“ BestÂ” SJT responses 1 Tett et al. (1991) argued that this was because Barrick and Mount neglected to consider absolute values in their analyses, effectively causing significant negative and positive values to Â“cancel each other out.Â” Barrick, Mount and others continue to assert that Tett et al. Â’s analytic methods were mathematically incorrect.
19 reflect an individualÂ’s typical behavioral preferences (which should be a function of personality) and are most likely to be relevant in highly autonomous situations. Similarly, Â“WorstÂ” responses would be most relevant to situations in which individualsÂ’ personalities have less influence on their behavior and they are not able to support the behaviors they would consider Â“BestÂ”. Cucina, Vasilopoulos and Leaman (2003) also found higher correlations between personality and situational ju dgment when using measures of narrower constructs. Clearly those personality factors that play a role in situational judgment response require further empirical clarification. Another construct relevant to management that calls for additional research i s leadership. Leadership Why Leadership? Situational Judgment tests are tools that are most often used to select managers and supervisors (Hanson & Ramos, 1996). It is therefore important to consider how characteristics of leadership may affect SJT resp onses. Leadership is defined as the process of influencing other group members to achieve organizational goals (Greenberg & Baron, 1997). Organizations may or may not make a distinction between leaders and managers: leaders are responsible for the vision of the company while managers are responsible for the implementation of that vision (Greenberg & Baron, 1997). This distinction seems somewhat arbitrary: although some management employees do not play a role in setting organizational goals, a manager is often perceived as the leader of his or her work group and must contribute as a leader on a smaller scale. This is especially true
20 when the organization is so large that employees have little to no contact with the leader of the company, when the organiza tion is so small that leaders must also take on the duties of management, or in organizations that encourage participation in goal setting and decision making from all levels of the company. Numerous theories of leadership have been set forth over the las t 100 years of research (Yukl, 1994). Leadership theory began with the idea that great leaders are born, not made, called the Â“great man theoryÂ” Â– that particular traits set leaders apart from ordinary people (Locke, 1991). Although most researchers have dismissed the Â“great man theory,Â” support for a solely trait (or dispositional) theory of leadership continues in recent literature (e.g., House, Shane & Herold, 1996). Current trends suggest that leader traits are important, but that successful leadersh ip is also influenced by situational variables. Called contingency theories, these models suggest that leaders must adapt their behavior based on the organizational environment (Schriescheim, Tepper & Terault, 1994). Modern contingency leadership theorie s also typically address the relationship between leaders and their subordinates (Yukl, 1994). One of the more prominent modern contingency theories of leadership is Path Goal theory. Path Goal Theory of Leadership The Path Goal theory of lead ership addresses how formal supervisors affect the motivation and satisfaction of their subordinates (House, 1996). It is a dyadic theory, meaning that it concerns itself with individual relationships between a supervisor and each subordinate (House, 1996 ). When House first proposed Path Goal theory (1971), he was concerned only with establishing that supervisors served the needs of subordinates in
21 two different ways: supervisors 1) helped subordinates to follow a course toward an outcome (the Â“path goalÂ” component), and 2) satisfied subordinate needs. He adopted the constructs of Consideration and Initiating Structure that were originally advanced by the Ohio State Leadership Studies (Stogdill & Coons, 1957). Consideration is defined as the amount of co ncern and empathy a leader shows toward subordinates (Judge, Piccolo & Ilies, 2004). Initiating Structure refers to how much a leader defines roles for self and subordinates, sets goals, and is achievement oriented (House, 1996). Not surprisingly, Consid eration has been consistently shown to correlate more strongly with subordinate satisfaction, while Initiating Structure has been shown to correlate more strongly with performance and effectiveness (Judge, Piccolo & Ilies, 2004). One major drawback of thi s two factor leadership theory is that Consideration and Initiating Structure were initially supposed to be orthogonal, but numerous studies have shown that there is a significant correlation between them (Judge, Piccolo & Ilies, 2004; Fleishman, 1995; Bas s, 1990). Later, House and Mitchell (1974) modified Path Goal theory into the traditional four factor model that is known today. Directive leadership is described as providing structure and expectations to subordinates; essentially, telling subordinates w hat they are supposed to do. Participative leadership is described as consulting with subordinates and taking their opinions and suggestions into account when making decisions; in effect, this approach allows subordinates to help decide what is to be done Both Directive and Participative leadership are components of path goal clarifying behavior (Evans, 1996). Supportive leadership is described as creating a supportive work environment and being
22 concerned for the welfare of subordinates. Achievement or iented leadership is described as setting challenging goals and emphasizing performance excellence. Both Supportive and Achievement Oriented leadership are components of satisfying subordinate needs (Evans, 1996). It is important to stress that although these characteristics would be demonstrated with marked behavioral contrasts, they are not polar opposites and are not mutually exclusive Â– it is possible for a leader to be both Supportive and Achievement Oriented, for example (Greenberg & Baron, 1997). Further, Path Goal theory also states that it is up to the leader to motivate subordinates through appropriate use of the components described above and that the effectiveness of this motivation is contingent on the degree of structure present in the work being performed (House, 1996). House later reformulated his own theory again, using Path Goal theory as a springboard for Charismatic Leadership theory and most recently, the Path Goal theory of Work Unit Leadership (House, 1996). Path Goal Theory in Res earch Path Goal theory has enjoyed moderate support in empirical research. A meta analysis performed by Judge, Piccolo and Ilies (2004) cited numerous studies that found little to no relationship between contingency theories of leadership and outcome vari ables. However, the same study analyzed several hundred correlations between consideration/initiating structure factors (the bases for Path Goal theory) and leadership outcomes such as subordinate satisfaction and leader job performance, and reported
23 mode rate average correlations for both of the above factors with numerous leadership outcomes (Consideration .48, Initiating Structure .29; Judge, Piccolo & Ilies, 2004). A handful of recent studies have shown that specific elements of the Path Goal model (par ticularly participative leadership) influence subordinate outcomes. Coleman (2004) determined that managers with more cooperative beliefs and ideals about organizational power relations were more likely to engage in participative leadership behavior than those with more competitive beliefs and ideals about organizational power relations. This effect was enhanced by the use of subliminal priming (quickly showing words related to competitive or cooperative beliefs on a computer screen) so that competitive p riming reduced the participative leadership behaviors of even those with more cooperative beliefs about organizational power. Oshagbemi (2004) demonstrated that older managers tend to use significantly more participative leadership behaviors than young er mangers. However, differences in the amount of directive leadership behaviors between older and younger managers were not statistically significant. Somech (2003) studied demographic differences between leaders and subordinates and showed that differ ences in age, gender, and level of education between the leaders and subordinates decreased the amount of participative leadership that was exhibited. These effects diminished over time, with the exception of dissimilar genders, which intensified over tim e. In other words, the longer a demographically dissimilar leader and subordinate worked together, the more likely the leader would display participative leadership, except in the case that the leader and subordinate were of
24 opposite genders; in this case the leader was less likely to display participative leadership after a longer relationship. Kahai, Sosik and Avolio (1997) showed that participative and directive leadership provided diverse results from subordinates in an electronic meeting situation. Under participative leadership conditions, subordinates provided more suggested solutions, were more supportive, and were less critical of the situation, than under directive conditions. There is no research that considers the effects of Path Goal charact eristics on situational judgment responses. However, Somech (2003) suggested that the relationship between participative leadership and organizational culture and structure should be explored in future research. Path Goal theory is a useful way to think about leadership from the perspective of organizational culture. Path Goal theory was not used in this research to predict subordinate performance at different levels of task structure. Instead, the four characteristics identified by Path Goal theory wer e used to create fictitious leadership cultures to demonstrate how organizational leadership culture can affect situational judgment responses. Specifically, it is likely that respondents will attempt to display congruence between their reported behavior (measured by responses to a situational judgment test) and the fictitious organizational leadership cultures that were presented in this research. The desire to display such congruence is due to social desirability, or more specifically, impression manage ment, or the ability to Â“fake good.Â”
25 Impression Management and Faking What are Impression Management and Faking? Impression Management is the attempt to present a positive impression of oneself to someone else (Ones & Vishwesvaran, 1998). A Â“posi tive impression,Â” in the context of selection, refers to the representation that best fits into organizational norms; Â“someone elseÂ” refers to the person making the selection decision. There are numerous studies that show how impression management is empl oyed during selection interviews ( e.g. Kristof Brown et al. 2002; Silvester et al. 2002; Vishwesvaran et al. 2001). It has similarly been written that responses to personality assessments are self presentations, not self reports (Hogan, 1998). That i s, the response that a person provides to an instrument, especially when the person is highly motivated to succeed on the instrument (such as a candidate for employment), is likely to reflect how the person would like to be seen, rather than how they truly are. It is likely the inference can be extrapolated to situational judgment tests: respondents are likely to choose the behavior that they perceive is most acceptable to the organization, rather than the behavior they would truly exhibit. This type of i ntentional distortion is better known as faking. There is good evidence from past research that participants are motivated to Â“fake goodÂ” even when they are asked to do so only for research purposes ( e.g. McFarland & Ryan, 2000). Likewise, individuals c an fake selectively, that is, on only certain parts of selection instruments (Dalen, Stanton & Roberts, 2001). The same research showed that the amount of information presented to an individual has little influence on how much they choose to fake (Dalen, Stanton & Roberts, 2001).
26 Situational judgment tests are about judgment; thus they appear to tap an ability that is expected to be relatively enduring. As previously mentioned, such tests typically correlate with measures of cognitive ability. The natu re of the questions also appears to tap maximal rather than typical performance, especially when the stem is written to address Best and Worst responses to the situation (Hooper, Cullen & Sackett, 2006). However, many SJTs (including the one used in the c urrent research) concern issues of social conflict. Therefore, there is good reason to believe that individual differences outside of analytical judgment, such as beliefs regarding norms, customs, personal values and experience, personality, and ideas abo ut the testing organizationÂ’s values may all influence the choice of the best and worst response to SJT items. In fact, this research expected that all motivated respondents would respond to a situational judgment test with some degree of impression manag ement or faking, limited in only two circumstances: when extreme differences between respondentsÂ’ individual values and the organizational values presented create cognitive dissonance, or when not enough information about organizational values is presented for respondents to form any impression. These circumstances were controlled in the study through the use of manipulation checks. Experience What is Experience? When industrial and organizational psychologists refer to Â“experience,Â” they usually mean a simple measure of time on the job, often the sum of all time spent on similar jobs with different organizations. A seminal meta analysis performed by Quinones, Ford and Teachout (1995), showed that out of 22 studies of experience
27 completed in the previ ous 20 years, all but 2 used time on the job or time with the company as their experience measures. Further, the analyses for 15 out of the 22 studies were computed at the level of job tenure (7 were computed at the level of organizational tenure). The variable Â“experienceÂ” is a common way to operationalize job knowledge. More experienced workers are typically expected to perform at higher levels, make fewer mistakes, and require less supervision (Greguras, 2005). Popular belief suggests that job knowl edge is directly related to time on the job, although it appears that experience may be asymptotically related to job knowledge (McDaniel, Schmidt, & Hunter, 1988), or even parabolically related to job knowledge (Sturman, 2003). That is, job knowledge doe s not continue to increase with time on the job, but either reaches asymptote after a certain amount of time (when little or no job knowledge and skills remain to be learned), or begins to decrease after a certain amount of time (when job knowledge and ski lls are no longer state of the art). The word Â“experienceÂ” usually refers to job experience, and thus job knowledge; however, there are different kinds of work related experience that an individual can have. For example, time with one particular organiza tion, even across numerous and unrelated jobs, could be an operationalization of organizational experience (or Â“organizational socialization,Â” see Quinones, Ford and Teachout, 1995). Organizational experience may be related to what Sternberg (1995) referr ed to as tacit knowledge: the individual understands organizational rules, and formal and informal procedures, even if the Â“rulesÂ” are never explicitly stated (Sturman, 2003). Management in companies that promote
28 from within can be assumed to have a moder ate to high level of organizational experience, since workers would be promoted to management after some years of service. On the other hand, managers who are recruited externally will probably have a low level of organizational experience when they are n ew to their management positions. These externally recruited managers may have more trouble adjusting and performing than those with more organizational experience. Similarly, time in a particular type of organizational culture, even across numerous and unrelated organizations, might be considered culture experience. That is, the individual understands what broad goals, values, and ideals are emphasized, on both formal and informal levels. It follows, as above, that when new management is sought, manage rs with highly incongruous cultural experience in their previous organizations will have more trouble performing and adjusting than those managers with more congruous cultural experience. Although there is very little research on the aforementioned types of experience, it was hypothesized that at least job experience may have important effects on SJT responses. Experience as Antecedent to Performance Job experience (and thus job knowledge) is one of the most widely recognized predictors of job performan ce (Kolz, McFarland, & Silverman, 1998). McDaniel, Schmidt and Hunter (1988) found a mean correlation of .32 between job experience and performance in their meta analysis across multiple occupations.
29 Experience and SJTs SJTs are highly related to j ob knowledge (and thus experience), which presents a problem in terms of testing inexperienced respondents who may have little or no experience in the situations as presented (Weekley & Jones, 1999). This is likely more of a concern on a SJT of technical, or Â“hardÂ” skills, which are typically very job specific, than with SJTs of Â“softÂ” skills (such as dealing with subordinates) that may be commonly used across numerous jobs. Quantifying Experience It is difficult to assign value to experience in terms of months or years in a position. Several theories of expertise do exist. Most commonly, experience is broken down into three to five linear stages proceeding from novice to expert ( e.g. Anderson, as cited in Genberg, 1992; Dreyfus & Dreyfus, 1985). Howev er, developing expertise may not be a linear process. Although job experience and organizational experience are highly correlated with time on the job, little research exists on how much time at work separates Â“novicesÂ” from Â“expertsÂ” (Genberg, 1992). Da ley (1999) states, Â“competent professionals have usually been in practice three to five yearsÂ” (pp. 134 135). It must be noted that this research is focused on expertise for nurses, and what Daley calls a Â“competent professionalÂ” would fall about in the m iddle of the linear novice to expert model proposed by Dreyfus and Dreyfus (1985). It is difficult to find recommendations on quantifying expertise for managers, in part because the duties of a manager can be quite different across many different organiza tions. For the purposes of this research,
30 both job and organizational experience were categorized into five distinct groups based roughly on Dreyfus and DreyfusÂ’ (1985) model. Research Hypotheses The nomological network that follows in Figure 1 is pre sented for the purposes of hypothesis testing only. The model is intended to illustrate overall hypothetical relationships between the constructs described above and to show that they may affect responses to situational judgment tests. It is not intended as a structural equation model. Figure 1 : A Nomological Network of Constructs Affecting SJT Response
31 This model suggests that Personality and Experiential constructs, in addition to attempts to impression manage, affect responses to s ituational judgment tests. The model also demonstrates the contribution of organizational leadership culture, and how it can indirectly influence the effects of Agreeableness. The Experiential portion of the model shows how job experience may be thought of as nested within organizational experience, which is similarly nested within culture experience. Job experience is shaded to indicate the dominance of this construct in the literature on experience. However, the model in Figure 1 argues that both orga nizational and culture experience may also contribute individually to SJT responses. Finally, it is realistic to assume that there are other, unexplored constructs that play a role (and that may affect the Personal and Experiential constructs), either on an individual or an organizational level. For example, as stated above, there is ample evidence that cognitive ability can influence responses to SJT items. Such constructs are beyond the scope of this research; their presence is included only to demonst rate that the model is not a complete explication of what factors affect situational judgment responses. The following hypotheses are the central questions of this research. More complex hypotheses are illustrated graphically to clarify the expected e ffect. The Effects of Agreeableness on SJT Scores Hypothesis 1: SJT scores will be affected by the personality variable of Agreeableness regardless of leadership culture manipulation. That is, respondents with a relatively higher level of Agreeableness wi ll choose different responses to situational judgment
32 items than those with a relatively lower level of Agreeableness, regardless of which organizational leadership culture is presented. The Effects of Organizational Leadership Culture Manipulations on SJT Responses Hypothesis 2: Responses to situational judgment items are influenced by the targetÂ’s perception of the leadership culture of the organization. Corollary 2a: A motivated test taker who has knowledge of the testing organizationÂ’s leadership cult ure will choose those responses that best reflect the test takerÂ’s understanding of the leadership culture so that scores will differ between strong Participative/Supportive and strong Directive/Achieving cultures. Corollary 2b: A motivated test taker wh o is presented with no information about the leadership culture of the testing organization will choose responses without being influenced by leadership culture so that scores will differ between a Neutral condition and the strong conditions described abov e. Corollary 2c: Item level responses will differ across disparate leadership cultures. The Interaction of Agreeableness and Organizational Leadership Culture Manipulations on SJT Scores Hypothesis 3: The relationship between Agreeableness and SJT scores w ill be different in discrete leadership culture conditions such that High Agreeableness shows a positive relationship with SJT responses in some conditions (Participative/Supportive and Control), and a negative relationship with SJT responses in other cond itions (Directive/Achieving), as indicated in Figure 2.
33 Figure 2: Agreeableness X Leadership Characteristics Interaction Effects on SJT Scores Agreeableness X Leadership Culture Interaction Lo Agree Hi Agree Agreeableness SJT Score Control Control Directive/ Achieving Participative/ Supportive Participative/ Supportive Directive/ Achieving The Effects of Experience on SJT Scores Hypothesis 4: SJT scores are influenced b y Job Experience such that scores on an SJT will be higher for experienced vs. inexperienced participants, regardless of organizational manipulation. Essentially, experienced managers likely have a deeper understanding of how to respond to the different de mands of discrete organizations and organizational cultures, and will make judgments appropriately. The Interaction of Job Experience and Organizational Leadership Culture Manipulations on SJT Scores Hypothesis 5: SJT scores under different organizat ional manipulations will be moderated by job experience. That is, when a relatively low degree of job experience is present, information on organizational leadership culture may have a smaller effect on judgment. This hypothesis may hold especially true i n the Directive/Achieving condition.
34 Inexperienced participants will likely not attach as much significance to the cultural information presented in the manipulation as experienced participants, and will thus score lower in the Directive/Achieving conditi on. The same effect may also occur in the Participative/Supportive condition, but it is not expected to be as pronounced (please refer to Figure 3). Figure 3: Experience X Leadership Characteristics Interaction on SJT Response Experience X Leadership Culture Interaction Lo Experience Hi Experience Experience SJT Score Control Control Directive/ Achieving Participative/ Supportive Participative/ Supportive Directive/ Achieving Method Participants Two samples were recruited for the experimental study. The first sample consisted of retail sales managers and assistant retail sales managers recruited from a major national telecommunications company. Retail sales managers are line employees who are responsible (within this company) for day to day operations of retail stores, where cellular handsets and peripherals are sold, customer accounts are processed, and customer service issues are handled on a person to person basis. A ssistant retail sales
35 managers have very similar duties and differ from Â“fullÂ” managers only in their level of experience or tenure with the company. Retail stores usually have a single retail sales manager or a single assistant retail sales manager; larg er stores may have a retail sales manager and an assistant retail sales manager. Retail sales managers are typically responsible for an individual retail location of 8 10 employees (Jonathan Canger, Associate Director of Staffing and Talent Acquisition, O ctober 4, 2006, personal communication ). District Managers throughout the company (226) were asked to provide the names of five retail sales managers or assistant retail sales managers within their districts to participate in this research. Approximately 149 District Managers responded (response rate = 66%), resulting in a list of 745 potential participants. Study information was emailed to the participants with a link to participate in the research online. A total of 386 managers and assistant managers responded to the request to participate, and 258 completed the entire research instrument (overall response rate = 35%). The sample consisted predominantly of retail managers (N of Managers = 229, or 89%; N of Assistant Managers = 23, or 9%, 6 employees, or 2% did not respond to this item). Demographic data on experience for this sample is presented in Table 3. The second sample in the main study consisted of 138 undergraduates from a large southeastern university. Students from introductory psycholog y courses volunteered to participate in the research in partial fulfillment of course experimental participation requirements. Although data on major was not collected, because introductory psychology courses are core requirements, the students within lik ely represented a diverse mixture of backgrounds and potential majors ( i.e., not all the
36 students were psychology majors). Student data were controlled for any part or full time management experience. Demographic data on experience for this sample are p resented in Table 3. Table 3 : Comparative Experience for Managers vs. Students Managers Students Management Experience Frequency Percent Frequency Percent 0 years (None) 0 0% 70 50.7% Less than 1 year total 10 3.9% 22 15.9% 1 to 3 years total 43 16. 7% 26 18.8% 3 to 5 years total 38 14.8% 10 7.2% More than 5 years total 166 64.6% 10 7.2% Total N 257* 138 Standard Deviation 0.91 1.28 *1 professional participant did not respond to the experience items Procedure Each participant responded to a seri es of situational judgment items as described in the introduction to this research (see Appendix B). Three separate sets of instructions were developed for both the student sample and the management sample. Each set of instructions consisted of a descrip tion of a hypothetical organization and a fabricated email about the organization from a fictional acquaintance within that organization. All participants responded to the same situational judgment items after being exposed to one of the three conditions: 1) The information provided in the Â“Participative/SupportiveÂ” condition suggested that the primary goal of management in this particular organization is to be highly
37 supportive of subordinates even at the expense of profitability (see Appendix A). For examp le, subordinates are encouraged to contribute to and correct their supervisors, while supervisors are expected to provide a high amount of job related and personal support to their subordinates. 2) The information provided in the Â“Directive/AchievingÂ” condi tion suggested that the primary goal of management in this particular organization is to focus on profitability even at the expense of subordinate support (see Appendix A). That is, subordinates are discouraged from contributing to or correcting their sup ervisors, while supervisors are instructed place organizational goals ahead of the personal and professional needs of subordinates. 3) No information was provided in the Â“ControlÂ” condition so that respondents were able to draw their own conclusions without any kind of direct influence (see Appendix A). Participants were assigned to one of the three manipulation conditions based on their date of birth. The first item on the survey asked participants to report their date of birth (day of the month only), and program logic took participants immediately to the correct manipulation condition, then on to the main survey. The manipulations were evaluated in a brief pilot study to ensure that they were being interpreted by participants in a manner consistent with P ath Goal Theory and the intent of the research. Please refer to the section of this dissertation entitled Â“Preliminary Analyses: Pilot Testing the Manipulations of Organizational Leadership CultureÂ” in the Â“ResultsÂ” section for supporting information.
38 Par ticipants were asked to imagine they are applying for a job that they are very motivated to get, the idea being that they would therefore try their best to model response sets that would be acceptable to the company to which they are applying. Upon complet ing the situational judgment measure (Appendix B), participants were asked to complete a secondary inventory (see Appendix C) consisting of personality items, experience items, and manipulation checks (to determine whether they were dissimulating as instru cted during the situational judgment portion of the research). Participants were instructed that they should now answer honestly and no longer attempt to fit their responses to the information they read at the beginning of the research. All participants responded to the research measures electronically. Participants were emailed a web address (URL) to access an online Â“survey hostingÂ” website that displayed one of the instructional manipulations followed by the research instruments described. Data was d ownloaded from the websiteÂ’s database after collection, and coded and analyzed using Microsoft Excel and SPSS 11.5. Measures Situational Judgment Inventory: The SJT 280 Although managers certainly have additional job responsibilities, it is argued that on e of the most important and characteristic duties of a manager is to supervise subordinates. Therefore, the primary measure was a set of situational judgment items that all address dealing with subordinates (see Appendix B). These items were derived from several sources. A total of 26 situational judgment items were included. This author created nine items specifically for use in this research. Eight items were taken
39 from the ProveIt Manager by Kenexa, Inc. Two were taken from the Supervisory Skills Inventory (SSIÂ™) by gNeil, Inc. Seven were based on items from an inventory created by the Personnel Decisions Research Institute for non commissioned officers in the U.S. Army [because this measure was specifically created for Army officers, wording was e dited to make the items more appropriate for a corporate setting. The nature of the items was not changed] (Hanson & Borman, 1992). All items are used in this research with the permission of the test development companies (Kenexa, Inc., gNeil, Inc., and the Personnel Decisions Research Institute). All three instruments are highly researched selection instruments that include situational judgment items on numerous management related topics; with the exception of the PDRI measure, these tests are commercia lly available for private use. Effectiveness of a particular response was expected to be subjective and highly culture and experience dependent. Therefore, six separate scoring keys were initially created for each of the three conditions described abov e, with both an experienced and a novice sample. This was necessary because what constitutes effective performance is thought to be different in organizations with disparate leadership cultures like those described above. Further, level of experience was expected to contribute to interpreting the effectiveness of responses. Scoring keys were developed through a pilot study of novice and experienced raters. Undergraduate students at a major southeastern university created the novice scoring keys for the SJT Inventory by rating the Effectiveness of every response choice on a 1 4 scale (Note that this rating procedure was more exhaustive than the Most/Least
40 Effective rating procedure used in the main study). Upper level management at a major southeastern freight and shipping company created the experienced scoring keys in similar fashion. The experienced (Manager) and novice (Student) ratings were converted to analogous scores and analyzed to demonstrate the item difficulty and reliability of the instrum ent. Although alphas for the Manager items were acceptable (range of = .80 .85), the alphas for two of the three Student keys were slightly lower than the standard of .70 (Participative = .78; Directive = .66; Control = .64). Further, all keys showed numerous negative item total correlations that were difficult to interpret. The finalized scoring keys were created to evaluate the responses of participants in each condition; however, due to potential reliability issues and other concerns descr ibed later, an additional keying method was developed. For supporting information on the development and abandonment of the initial scoring keys, please refer to the portion of this dissertation entitled Â“Preliminary Analyses: Pilot DataÂ” in the Â“ResultsÂ” section. Participants in the main research sample were asked to choose one Most Effective and one Least Effective response choice for each item (see Appendix B). This response format was chosen because it is more likely to address maximal performance (al so referred to as Â“should doÂ” or Â“could doÂ” performance), which is what would be expected from a job applicant (Ployhart & Ehrhart, 2003). Furthermore, it was expected that responding with Most Effective and Least Effective would avoid any cognitive disso nance that might result from asking participants to choose the responses that are Â“Most LikeÂ” and Â“Least LikeÂ” them, due to the likelihood that participants are
41 dissimulating. That is, since participants in some conditions were expected to impression mana ge their responses, it was expected that separating the participant from the response might lead to answers that were better tailored to fit the manipulation. The revised key method assigned scores to participants based on frequency of response endorsemen t. Each participant was assigned two scores for each item: one score for the response endorsed as Most Effective and one score for the response endorsed as Least Effective. The scores were equal to the proportion of all experimental participants (from bo th Student and Manager samples) who endorsed those responses. The frequency based key method is described further in the section entitled, Â“Preliminary Analysis: Creating a New Frequency Based Key Using Experimental DataÂ” in the Results. The frequency ba sed method allowed for all three conditions and both levels of experience to be scored on the same key while still retaining individual differences in response patterns. This in turn allowed for more sophisticated statistical analyses such as analyses of variance and covariance to be performed on the whole dataset. Secondary Inventory: The SPE 30 A secondary inventory was used to measure the Big 5 personality factor of Agreeableness, which was expected to moderate responses to the Situational Judgment Items. Ten items that measure agreeableness were taken from the IPIP website of public domain test items available for research purposes ( http://www.ipip.ori.org ). These items relate to Agreeableness, and address Costa, McCrae and DyeÂ’s (1991) facets of Altruism and Tender mindedness. However, four additional items were written by this author to address the other Agreeableness facets (Trust, Straightforwardness, Compliance, and
42 Modesty) that appear be related to the experimental construct of dealing with subordinates, but that did not appear to be addressed by the IPIP items. The secondary inventory included demographic items about management interest and experience, as well as a question about category of industry for the student sample. The industry ca tegories for this question were based on the U.S. Department of Labor Bureau of Labor Statistics Standard Occupational Codes (SOC), taken from the BLS website ( http://www.bls.gov ). Categories that were not likely to include management positions, and the c ategory Â“Management OccupationsÂ” were excluded. These items were used to control for management experience and interest, specifically in the undergraduate sample. These items also included a categorical variable of management experience for the incumbent manager sample. Based on the work of Dreyfus and Dreyfus (1985) described earlier, the experience variable included five distinct categories. Management experience on the part of the students was compared to management experience for the professionals to ensure that there were significant differences between the two experience conditions; this analysis is presented in the Â“ResultsÂ” section of this research. Finally, this inventory included manipulation check items to ensure that participants are respon ding within the provisions of the instructional manipulations of organizational culture. The items on this inventory were scored on a 5 point Likert scale, with response options ranging from Â“I Very Much DisagreeÂ” to Â“I Very Much AgreeÂ”. The secondary in ventory is included as Appendix C.
43 Results Preliminary Analysis: Pilot Testing the Manipulations of Organizational Leadership Culture Before the manipulations were ever presented to the pilot or experimental samples, it was important to ensure that the participants would interpret the manipulations as intended by the experimenter. A short scale was developed to measure reactions to the manipulations. The Manipulation Pilot Inventory (see Appendix D) included 9 items based on characteristics of leaders hip styles as proposed by Path Goal Theory (House & Mitchell, 1974). A total of 14 undergraduate students were exposed to one of the three manipulation conditions, and then asked to evaluate whether the company in the description that they read matched im portant characteristics of Participative or Directive leadership. It was expected that there would be significant differences in how participants judged each manipulation condition based on leadership characteristics. A one way ANOVA with accompanying po st hoc Tukey tests were computed, and results for individual leadership characteristics appear in Table 4. Table 4: Interpretations of Leadership Characteristics Between Manipulation Conditions EmployeesÂ… LeadersÂ… Condition Are Told What is Expected Are Part of Decisions Are Asked for Suggestions Are Approachable Are Concerned Have High Performance Expectations Participative 4.20 (a) 5.00 (a) 4.80 (a) 4.60 (a) 4.60 (a) 4.60 (a) Directive 4.80 (a) 1.40 (b) 1.80 (b) 1.60 (b) 1.80 (b) 4.80 (a) Neutral 3.25 (b) 4.00 (c ) 4.00 (a) 4.00 (a) 3.75 (a) 3.00 (b) Conditions that were significantly different ( p < .05) are designated by different letters (a, b, c); mean response to each characteristic on a 1 5 scale is presented in boldface These data suggest that participants ar e likely to interpret the manipulation conditions as different on numerous leadership characteristics proposed by Path Goal
44 Theory. Responses to two leadership characteristics items were not significantly different, although the effects were in the expect ed direction: Â“Â…employees are told how to performÂ…Â” and Â“Â…leaders set challenging goalsÂ…Â” suggesting that these two characteristics were not interpreted as typical of one particular leadership style, or that this characteristic was not addressed strongly e nough in the manipulation. However, the significant differences observed are a close fit with the experimental purpose of the manipulations. Finally, responses to the item on the inventory that addressed whether participants would like to work in the env ironment described were not significantly different across conditions. This is meaningful because it suggests that the environment described in the Directive condition was not seen as objectionable, potentially decreasing response bias from participants i n that group. Preliminary Analysis: Pilot Data Upper level managers from a major southeastern freight and shipping company and undergraduate students from a major southeastern university were recruited to pilot the situational judgment instrument and to d evelop a scoring key (see Appendix E). These data were obtained chiefly to determine that the situational judgment items included in this research had an acceptable level of reliability. A total of 118 participants (30 across the three professional/manag er conditions, 88 across the three novice/student conditions) were used to create six unique keys. Each participant viewed one of the instructional manipulations described previously before responding to the situational judgment items. Participants were informed that their responses would be used to create a key for a new situational judgment test, and asked to rate every response choice on a 1
45 4 scale (where 1 = A Very Effective Response, 2 = A Somewhat Effective Response, 3 = A Somewhat Ineffective Resp onse and 4 = A Very Ineffective Response). To create the keys and to measure reliability, scores for each response choice were averaged across all participants in that condition. These means were summed to create an item level score for each individual b ased on the response choices they designated as Very Effective minus the response choices they indicated were Very Ineffective (since participants could designate multiple response choices as Very Effective or Very Ineffective, item level scores were calcu lated based on the following formula: means very effective /N means Â– means very ineffective /N means). This method allowed for the creation of analogous scores across all three conditions and both levels of experience. To avoid the likelihood that the range of responses and thus varianc e would be different across the different levels of the independent variables, every participant was scored on each of the six keys and those six scores were added together to create a composite score from all six keys (referred to in this research as the Summed Six Key Method). However, at this point, a minor setback in data collection occurred. Preliminary Analysis: Creating a New Frequency Based Key Using Experimental Data The professional sample in the pilot study, taken from the freight and shipping company became unavailable at the completion of pilot data collection, and was not available to participate in the experimental research. Therefore, because of the small sample size of the pilot group, because of an unusual number of negative item total correlations, and to ensure the appropriateness of keyed responses across organizations, a new key was created based on the response frequencies of the experimental data itself.
46 Each participant was given two scores for each item: one score for the respon se he or she endorsed as Most Effective and one score for the response he or she endorsed as Least Effective. The scores were equal to the proportion of all experimental participants (both Student and Manager samples) who endorsed those responses. For ex ample, if Participant X chose Response Â“aÂ” as Most Effective for Item 1, and Response Â“aÂ” was chosen by 38.6% of all respondents, Participant X would receive a score of .386 for Most Effective response for Item 1. Most Effective and Least Effective scores were summed for each item and then item scores were added together to create a composite score for each participant on the entire instrument. The following formula illustrates how composite scores were created using this method: "( Most Effective freq + Le ast Effective freq ). This method allowed for all three conditions and both levels of experience to be scored on the same key while still retaining individual differences in response patterns. Using the experimental data to create the Frequency Based Key m ay be expected to result in multicollinearity and artificial inflation of scores compared to a new sample. However, use of this method of keying the data is justified because of the relatively large N (396), and because any differences observed between ex perimental conditions cannot be explained by intercorrelation. Although the maximum hypothetical score using this key is 52, the maximum obtainable score using this key was 30.2, the minimum obtainable score using this key was 2.9 (since every response ch oice that was endorsed at least once must have a non zero score and no response choice was endorsed at 100%). The alpha for the new, frequency based key was .76. A single item had a weak negative item total correlation, which did not strongly affect the reliability.
47 Therefore, all items were retained in the SJT. Finally, although the Six Key Method was not used in further analyses, it is interesting to note that scores using the Frequency Based Key Method and the Summed Six Key Method were highly correl ated [ r (395) = .954, p < .05]. Experimental Results: Manipulation Check Experimental participants responded to three items used as manipulation checks in the secondary inventory. These items were included to address whether participants answered different ly than normal because they were asked to act as though they really wanted the job, and whether the company description and email presented before the SJT provided clues about how to answer in order to get the job (the verbatim items are included in Append ix C). A composite score was created based on participantsÂ’ responses to these three items. A two way ANOVA demonstrated that while there was no significant difference between managersÂ’ and studentsÂ’ responses to these items, the responses from the Contr ol condition were significantly different from the responses from the Participative/Supportive and Directive/Achieving conditions [ F (2) = 10.99, p<.001]. The mean composites for the 3 manipulation check items are shown in Table 5. Table 5: Manipulation Ch eck Composite Scores Among 3 Conditions Condition Mean Standard Deviation N Participative/Supportive 9.06 2.36 114 Directive/Achieving 9.41 2.77 141 Control 7.90 2.31 141 These data suggest that the manipulations of Organizational Leadership Cultu re were working as intended; participants in the Control Condition were not as highly
48 influenced by the information presented as the other groups; that is, Control participants were more likely to answer as they normally would (instead of impression managi ng to get the job), and they were less influenced by the company description and email that was presented. Experimental Results: Hypothesis Testing A General Linear Model was created that included main effect terms for Agreeableness, the Leadership Cultu re manipulations, and Experience, as well as 2 way interactions between Agreeableness and the Leadership Culture manipulations and Experience and the Leadership Culture manipulations. For the convenience of the reader, the research hypotheses are restated below as the data are presented. Hypothesis 1 stated that scores on the situational judgment test would be affected by the personality variable of Agreeableness regardless of leadership culture manipulation. This hypothesis was tested by the signifi cance of the main effect for Agreeableness. A composite score was created for Agreeableness by summing responses to the 14 Agreeableness items on the secondary inventory (please refer to Appendix C). Reliability analyses of the Agreeableness scale were c onducted first to ensure that the Agreeableness items had acceptable internal consistency. Alpha was acceptable at .71 for the Agreeableness measure. Analyses of Variance were conducted to determine whether Agreeableness varied across Experience (Manager vs. Student) or Leadership (Participative/Supportive vs. Directive/Achieving vs. Control) conditions. No significant differences across conditions were found, suggesting that Agreeableness (as tested) varied consistently throughout the experimental sampl e. Since participants were assigned
49 to condition based on their date of birth, it is unlikely that any systematic variation occurred based on Experience or Leadership Culture Condition. Analysis of covariance (ANCOVA), using Experience Levels and the Lead ership Culture manipulations as fixed factors and the Agreeableness composite score as a covariate indicated a significant main effect for Agreeableness [ F (2, 387) = 4.162, p = .042] in predicting SJT scores. This analysis demonstrates differences in Effe ctiveness of responses. Means and standard deviations are shown in Table 6. Table 6: Means, Standard Deviations and Score Ranges for Agreeableness Mean Score Standard Dev. Observed Min Â– Max Potential Min Max 57.1 5.74 34 69 14 70 Discriminant anal ysis was also performed at a response level to determine if mean Agreeableness composite scores were significantly predictive of the choice of item response for each item. This analysis demonstrates differences in response choices regardless of Effectiven ess. A total of 7 out of 52 analyses were significant, or just over 13%. Hypothesis 1 was supported. Hypothesis 2 stated that scores on the SJT and item level responses on the situational judgment items are influenced by the targetÂ’s knowledge of the lea dership culture of the organization, as demonstrated by the presentation of distinct leadership culture manipulations (versus control). The first part of the hypothesis was tested by the significance of the main effect for the Leadership Culture manipulat ions [ F (2, 387) = 4.53, p = .011]. Post hoc Tukey tests were conducted on this result and demonstrated significant differences only between the Directive/Achieving condition and the Control condition. Descriptive statistics are presented in Table 7.
50 Table 7: SJT Score Means and Standard Deviations by Manipulation Leadership Manipulation N Mean Standard Deviation Participative/Supportive 114 23.34 2.87 Directive/Achieving 141 22.72 3.37 Control 141 23.91 2.83 Therefore, knowledge of Organizationa l Leadership Culture, notably in a Directive/Achieving culture may have a significant effect on SJT response. The hypothesis was tested at a response level by comparison of response frequencies through Chi square analysis. Chi square coefficients and Phi values were computed for each item to test whether the frequency distributions differed across the three culture manipulations for each item. Chi square coefficients were significant for 22 out of 52 analyses, or 42%, suggesting that item level responses differed between the culture conditions. Differences in item level responses among the manipulated conditions may suggest that participants tend to answer differently across conditions, regardless of whether their behaviors would be considered effective ( e.g ., different ineffective behaviors may be endorsed in different conditions). Significant Chi square and Phi values are presented in Appendix F. Hypothesis 2 was supported. Hypothesis 3 stated the relationship between Agreeableness and SJT scores are different in discrete leadership culture conditions such that High Agreeableness shows a positive correlation with SJT score in some conditions (Participative/Supportive and Control), and negative correlation with SJT responses in other conditions
51 (Directi ve/Achieving). This hypothesis was tested by examining the significance of the interaction term between Agreeableness and Culture. The ANCOVA showed no significant interaction between the Agreeableness covariate and Leadership Culture Condition [ F (2, 387) = 1.33, p = .266]. However, correlational analyses between Agreeableness composite scores and SJT scores demonstrated a small but significant positive correlation in the Participative/Supportive Condition, and no correlation in the Directive/Achieving an d Control Conditions. Means and standard deviations as well as correlations between Agreeableness and Leadership Culture Condition are shown in Table 8. Table 8: Means, Standard Deviations and Correlations Between Agreeableness & SJT Score by Condition Condition N Agreeableness Mean Score Agreeableness Standard Dev. Correlations Participative/Supportive 114 57.51 5.03 .191 ( p = .041) Directive/Achieving 141 56.27 6.26 .015 ( n.s. ) Control 141 57.59 5.69 .029 ( n.s. ) These small but potentially import ant differences in the relationship between Agreeableness Scores and SJT scores in specific Leadership Culture conditions are illustrated graphically below in Figure 4. However, Hypothesis 3 was not supported.
52 Figure 4 : Slopes of Agreeableness SJT Score Relationships by Leadership Culture Participative Control Directive 40 50 60 70 Agreeableness Score 10.00 15.00 20.00 25.00 Frequency Based Key Score b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b 40 50 60 70 Agreeableness Score b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b 40 50 60 70 Agreeableness Score b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b b Hypothesis 4 stated that SJT scores are influenced by Job Experience such that scores on an SJT would be higher for experienced vs. inexperienced participants, regardless of organizational manipulation. This hypothesis was tested by the significance of the main effect for Experience. First, it was important to be sure that the Manager and Student groups clearly differed in level of individual experience, based on the information presented in Table 3 abo ve. An independent samples t test was used to demonstrate that the experience level of the Management group was statistically different from that of the Student group [ t (393) = 21.1, p < .001]. The analysis of covariance showed a significant main effect for Experience [ F (2, 387) = 102.22, p < .001]. Descriptive statistics for the Experience variable are provided in Table 9. Hypothesis 4 was supported. Table 9: Mean SJT Scores and Standard Deviations by Experience Status Experience Status Mean SJT Score Standard Deviation Student 21.35 2.98 Manager 24.39 2.58
53 Hypothesis 5 stated that scores on the situational judgment test under different organizational manipulations would be moderated by job experience. This hypothesis was tested by examining th e significance of the interaction between Experience and the three Leadership Culture manipulations. The ANCOVA showed a significant interaction between Experience and Leadership Culture [ F (2, 387) = 3.804, p < .023]. Mean SJT scores and standard deviati ons are provided in Table 10. Boxplots provide a graphical representation of the interaction in Figure 5. Table 10: Mean SJT Scores and SDs by Experience and Leadership Condition Status Condition Mean SJT Score St. Dev. Participative/Supportive 21.93 3.02 Control 21.27 3.24 Student Directive/Achieving 20.97 2.69 Participative/Supportive 24.11 2.52 Control 25.15 1.45 Manager Directive/Achieving 23.79 3.30
54 Figure 5: Student & Manager SJT Scores Across Conditions 88 96 74 53 45 40 N = Leadership Culture Directive/Achieving Control Participative/Suppor Frequency Based Key Score 40 30 20 10 0 Experience Student Manager 333 312 384 320 395 376 375 183 72 111 83 113 58 295 272 277 117 119 7 1 This hypothesis was also tested at a response level by comparison of response frequencies through Chi square analysis. Chi square coefficients and Phi values were computed for each item within the Student and the Manager sample to t est whether the frequency distributions differed across the three culture manipulations for each item. Chi square coefficients were significant for 28 out of 104 analyses (52 Student items and 52 Manager items), or 27%, suggesting that item level response s differed between the culture conditions. Differences in item level responses among experience status and the manipulated conditions may suggest that participants tend to answer differently across experience status and condition, regardless of whether th eir behaviors would be considered effective ( e.g ., different ineffective behaviors may be endorsed in different
55 conditions). The tables of response frequencies are presented in Appendix G; Chi square and Phi values are presented in Appendices H and I. Hy pothesis 5 was supported. The analysis of covariance ( ANCOVA ) that was used to analyze the three main effects and the two interactions is included for reference as Table 11. Table 11: ANCOVA Model of Main Effects and Interactions Source Type III Sum of S quares Degrees of Freedom Mean Square F Sig. Corrected Model 982.98(a) 8 122.87 17.19 .000 Intercept 1409.07 1 1409.07 197.13 .000 Experience 730.67 1 730.67 102.22 .000 Leadership Culture 64.76 2 32.38 4.53 .011 Agreeableness 29.75 1 29.75 4.16 .042 Experience Leadership Culture 54.38 2 27.19 3.80 .023 Agreeableness Leadership Culture 18.97 2 9.49 1.33 .266 Error 2766.30 387 7.15 Total 219200.62 396 Corrected Total 3749.28 395 a R Squared = .262 (Adjusted R Squared = .247) Discussion This research was intended to be a preliminary step in critically examining unexplored constructs at both a personal and situational level that explain variance in responses to situational judgment tests, as recommended by earlier research (Pl oyhart & Weekley, 2006). There are limitless factors and constructs that may contribute to explained variance for this type of test, not only on a composite level, but also on an item level or even a response level. The construct of Experience has been w idely researched in the past, Agreeableness has received some attention; both were considered deserving of a second look in the context of SJT response. The Â“newÂ” construct of Organizational
56 Leadership Culture was included in this research because of this authorÂ’s first hand experience with commercial selection test publishers. It is not uncommon for commercial test publishers to base passing scores on Â“off the shelfÂ” or Â“cannedÂ” situational judgment tests (and other kinds of tests) on norms that may or m ay not be appropriate for every organization, especially due to differences in organizational cultures. This research suggests that Experience might play an important role in shaping responses to situational judgment tests. The result must be interpret ed with caution because there are variables that are confounded with the operational definition of Experience that were not controlled in this experiment ( e.g. age, education, and the organizational and cultural definitions of experience explained earlier in this research). The Organizational Leadership Culture and Agreeableness constructs may also provide a small contribution to SJT response. However, the constructs explored did not fit the model as expected, and undeniably the contributions of Leaders hip Culture and Agreeableness were small in comparison to the effect of Experience. One of the reasons that Organizational Leadership Culture may not have shown a robust significant effect is that participants may have seen the culture manipulations as tr ansparent. Those participants in the Control condition actually had the highest scores on the SJT, suggesting that maybe the manipulations caused participants in those conditions to carefully consider their way of thinking about the job. Alternatively, i t is possible that the Directive/Achieving condition was the most difficult of the three to interpret and
57 impression manage to fit, hence this condition showed lower mean scores than the other conditions, as put forward in Hypothesis 5. Agreeableness sho wed a small significant effect on SJT responses. This finding is in keeping with previous research and suggests that learning more about how personality traits are related to judgment will be useful in construct explication of situational judgment ( e.g. Motowidlo, Hooper & Jackson, 2006; McDaniel & Nguyen, 2001). This study suggests that high Agreeableness may be related to higher scores on situational judgment tests. Again, this result must be interpreted with caution, as it is also likely that Agreeab leness is perceived as a universally desirable personality trait among jobs that require considerable interpersonal interaction (Barrick & Mount, 2005). Further, the main effect for Agreeableness just barely achieved significance; this may be due to range restriction. There is a certain degree of confound between having an agreeable personality and asking participants to behave in a certain way. Asking participants to respond as though they are very interested in getting a job with a company described in a certain way may be easier for participants with a higher level of Agreeableness, regardless of the description. The ability of highly agreeable people to better impression manage regardless of Leadership Culture could explain why there was such a smal l effect for Agreeableness, as well as why there was no interaction effect between Agreeableness and the manipulations of Leadership Culture when highly Agreeable participants were initially expected to have lower SJT scores in the Directive/Achieving mani pulation, for example.
58 When the interaction between Experience and the manipulation of Leadership Culture was explored, significant influence on SJT response was observed. This is explained in part by the likelihood that job experience influences how well respondents Â“fitÂ” themselves into different aspects of organizational culture; those with more experience are likely to do what they have done in the past, regardless of the leadership culture of the organization, because they expect it to work. Alternat ively, those with less experience are more likely to see a need to fit in with implicit organizational policies to succeed. Limitations of this Research This research is a meaningful early step in construct explication, but it was not without its problems. One of the most obvious concerns is that access to the original management sample was lost after a small amount of pilot data was obtained. This loss was beyond the control of the experimenter and his colleagues at that organization, and is likely all too common when utilizing applied samples. It was fortunate that another Management sample became available, but it is clear that potential differences could exist between the organization used in the pilot research and the organization used i n the experimental research, while the student sample was taken from the same undergraduate population both times. While this change did not necessarily limit the validity of this research, it resulted in a rethinking of how to score participant responses after the study had been planned. Another concern was that the student experimental sample was small compared to the management experimental sample. This is unusual, as it is typically easier to recruit
59 students than professionals. However, having fewe r students resulted in unequal cell sizes, and may have adversely affected power. This is especially important in light of the fact that several significant relationships barely met the [ p < .05] convention. While the size of the student sample was parti ally under the control of this experimenter, a decision was made to limit student participation to a single semester to allow other researchers to take advantage of the universityÂ’s undergraduate research pool. Additionally, the management samples for b oth the pilot and experimental research were convenience samples. Participants in the managerial sample of the pilot study agreed to participate as a favor to this researcher and may therefore have been unfairly biased toward the research. Participation in the managerial sample of the experimental research was limited to those managers who deigned to respond to the voluntary research request. It is virtually impossible to report metrics for those managers who chose not to participate. Much of the data t ested showed a lack of homogeneity of variance (according to LeveneÂ’s test for Homogeneity of Variance). This presents some statistical concerns when interpreting any significant relationships. However, where possible, results were reported with unequal variances assumed in an attempt to correct for this finding. An obvious drawback to this research was the finding that numerous items on the SJT were answered consistently regardless of condition or experience. Psychometrically, this may suggest these it ems were too simple, or that particular sets of responses contained too many poor or transparent distracters. However, since many of these items were taken from actual situational judgment inventories that have been validated and
60 found to be reliable, thi s concern may be a detriment to the criterion validity of the instruments themselves as much as the research methods. The additional step of editing the instrument to remove or revise those items that were endorsed in the same way across all conditions/ex perience levels should be considered in replications of this research. Eliminating consistently endorsed items might show more significant effects within this research. Elimination or revision of these items could also be beneficial from an applied persp ective; although it would seem that items that are consistent across conditions would be useful to test developers, it is also possible that those items are not particularly predictive of organizational fit, or even performance across different organizatio ns. Directions for Future Research It is important that construct explication be continued to discover more about how and why situational judgment works in selection. This research could be advanced by performing content analysis of the items that wer e sensitive to the effects of Organizational Leadership Culture to determine if the items possess commonalities, and what distinguishes them from the items that were not significantly sensitive (per Appendices F, H and I). Likewise, a replication of this study would benefit from a higher degree of fidelity and realism if participants applying to actual organizations that clearly differ in their Organizational Leadership Cultures could be tested instead of relying on descriptions of fictitious organizations It would also be very interesting to perform similar research with different managerial positions and/or different industries. Many of the SJT items were originally written for staff management positions; that is, positions that have supervisory
61 respons ibilities over administrative personnel rather than those connected expressly to the product or service of the organization. The managerial sample that participated in the pilot research consisted entirely of staff managers, while the managerial sample th at participated in the experimental research consisted entirely of line managers. The implication is that managers who are involved in these two different functions may have very different perspectives about what to do in the same situation, based on dist inct differences in achievement and support orientations in staff vs. line managers (Church & Waclawski, 2001). A straightforward step toward better understanding the constructs behind SJT response would be to pilot an SJT on similar job titles in differe nt organizations (or different divisions or even teams within the same organization) to explore differential criterion related validity. It is possible that observed differences in the utility of the same SJT could be explained by situational or personali ty variables that are characteristically different in different organizations or micro organizations. Additionally, different aspects of leadership or management could be explored, rather than limiting the instrument to addressing Â“dealing with subord inates.Â” There continues to be a call for research on increased specificity in situational judgment tests; that is, focusing SJTs toward specific job tasks and abilities ( e.g. Ployhart & Weekley, 2006; McDaniel et al. 2001; Weekley & Jones, 1999). Alth ough dealing with subordinates was considered one of the most important job duties of managers in general, there are numerous other job duties that are likely equally important and equally common across all management jobs, for example conducting performan ce appraisals, handling escalations ( i.e. situations that are too demanding for subordinates to handle that are
62 consequently passed up to a manager), and dealing with internal or external customers. These and other job duties would likely show different effects in the context of Path Goal Theory; for example, the domain of conducting performance appraisals would be expected to be profoundly influenced by the Participative or Directive nature of an organization, while handling escalations might be more hea vily influenced by the Supportive or Achievement oriented nature of an organization. The use of Path Goal Theory was oversimplified for the purpose of limiting the manipulated conditions in this research. The combination of Participative with Supportive, and Directive with Achievement oriented characteristics were used to illustrate stereotypes of an employee friendly vs. an authoritarian, profit driven organization. It was expected that these stereotypes would be simplest for participant interpretation and impression formation. However, other combinations of the four Path Goal characteristics are entirely conceivable. Organizations might alternatively be considered Participative and Achievement oriented, or Directive and Supportive. These characterist ics could be combined in multiple ways in future research. It would be ideal to recruit participants from organizations that have decidedly different Leadership Cultures; this would render unnecessary the manipulations used in this study. Likew ise, future researchers should consider evaluating response choices on continua other than Â“effectivenessÂ”. For example, it would be useful and possibly meaningful to have pilot raters evaluate response choices for agreeableness, conscientiousness, or oth er personality variables to determine how important these
63 variables might be in determining what courses of action are effective. This follows from the implicit trait policy research of Motowidlo and colleagues (2006). Other ways that SJTs should be exam ined in the future include additional psychometric research, such as a comparative study that can demonstrate the most reliable and valid way to develop and score SJTs (Weekley, Ployhart & Holtz, 2006). For example, a larger expert pilot sample could be o btained so that the Conditional Key method could be effectively compared to the Frequency Key method. The future of situational judgment is bright because of the methodÂ’s low cost and high validity, as described above. As the U.S. becomes increasingly a service industry based nation, it is likely that SJTs will become even more useful in predicting performance because of their relationship with procedural knowledge (Ployhart & Weekley, 2006). That is, while work sample tests are highly predictive for jo bs requiring workers who are skilled in specific physical tasks (Schmidt & Hunter, 1998), many service jobs require abilities less objective in nature ( e.g. diagnosis, troubleshooting, creativity, etc. ) that are difficult to test. This research found sma ll, but significant differences based on a minute aspect of cultural differences (leadership culture) within a larger homogeneous culture (U.S. organizations). It follows that SJT research should take a global perspective in the near future, comparing sit uational judgment across regional and international cultures that are undoubtedly more diverse, possibly leading to richer differences in how judgment is engaged. This research demonstrates that the question of Â“why and howÂ” situational judgment tests wo rk likely requires a multifaceted answer. It is vital that future research
64 focus on additional personality and situational traits to learn more about what affects situational judgment. A better understanding of the constructs that influence SJT responses could ultimately lead to more effective tests that can better predict performance, turnover, job fit and other outcomes central to the field of employment testing.
65 References Barrick, M.R. & Mount, M.K. (1991). The Big 5 personality dimensions and jo b performance: A meta analysis. Personnel Psychology, 44, 1 26. Barrick, M.R., & Mount, M.K. (1993). Autonomy as a moderator of the relationships between the big five personality dimensions and job performance. Journal of Applied Psychology, 78 111 11 8. Barrick, M.R. & Mount, M.K. (1993). Yes, personality matters: Moving on to more important matters. Human Performance, 18, 349 372. Barrick, M.R., Mount, M.K. & Judge, T.A. (2001). Personality and performance at the beginning of the new millennium: Wh at do we know and where do we go next? International Journal of Selection and Assessment, 9, 9 30. Bass, B.M. (1990). Bass & StogdillÂ’s Handbook of Leadership New York: Free Press. Borman, W.C., Hanson, M.A. & Hedge, J.W. (1997). Personnel selection. Annual Review of Psychology, 48, 299 337. Callinan, M. & Robertson, I.T. (2000). Work sample testing. International Journal of Selection and Assessment, 8, 248 160. Church, A.H. & Waclawski, J. (2001). Hold the line: An examination of line vs. staff differences. Human Resource Management, 40, 21 34. Clause, C.S., Mullins, M.E., Nee, M.T., Pulakos, E., & Schmitt, N. (1998). Parallel test form development: A procedure for alternate predictors and an example. Personnel Psychology, 51, 193 208.
66 Cleveng er, J, Pereira, G.M., Wiechmann, D., Schmitt, N., & Harvey, V.S. (2001). Incremental validity of situational judgment tests. Journal of Applied Psychology, 86, 410 417. Coleman, P.T. (2004). Implicit theories of organizational power and priming effects on managerial power sharing decisions: An experimental study. Journal of Applied Social Psychology, 34, 297 321. Costa, McCrae, R.R. & Dye, D.A. (1991). Facet scales for Agreeableness and Conscientiousness: A revision of the NEO Personality Inventory. P ersonality and Individual Differences, 12, 887 898. Cucina, J.M., Vasilopoulos, N.L., & Leaman, J.A. (2003). The bandwith fidelity dilemma and situational judgment test validity Poster presented at the 18 th meeting of the Society of Industrial and Organ izational Psychology, Orlando, Florida. DÂ’Alessio, A.T. (1994). Predicting insurance agent turnover using a video based SJT. Journal of Business & Psychology, 9, 23 32. Dalen, L.H., Stanton, N.A., & Roberts, A.D. (2001). Faking personality questionnaire s in personnel selection. Journal of Management Development, 20, 729 742. Daley, B.J. (1999). Novice to expert: An exploration of how professionals learn. Adult Education Quarterly, 49, 133 147. Dreyfus, H. & Dreyfus, S. (1985). Mind Over Machine: The Power of Human Intuition and Expertise in the Era of the Computer New York: Free Press.
67 Evans, (1996). M.G. (1996). R.J. HouseÂ’s Â“A Path Goal theory of leadership effectiveness.Â” Leadership Quarterly, 7, 305 309. Fleishman, E.A. (1995). Consideration and structure: Another look at their role in leadership research. In F. Dansereau & F.J. Yammarino (Eds.), Leadership: The Multi level Approaches (pp. 51 60). Stamford, CT: JAI Press. Genberg, V. (1992). Patterns and organizing perspectives: A view of expertise. Teaching and Teacher Education, 8, 485 495. Greenberg, J. & Baron, R.A. (1997). Behavior in Organizations, 6 th Ed. Upper Saddle River, NJ: Prentice Hall. Greguras, G. J. (2005). Managerial experience and the measurement equivalence of perfor mance ratings. Journal of Business and Psychology, 19, 383 397. Hanson, M.A. & Borman, W.C. (1992). Development and construct validation of the Situational Judgment Test (Institute Report #230). Submitted to the U.S. Army Research Institute. Hanson, M .A. & Ramos, R.A. (1996). Situational judgment tests. In R. S. Barrett (Ed.) Fair Employment Strategies in Human Resource Management Westport, CT: Quorum. 119 124. Hogan, R. (1998). What is personality psychology? Psychological Inquiry, 9, 152 153. H ooper, A.C., Cullen, M.J. & Sackett, P.R. (2006). Operational threats to the use of SJTs: Faking, coaching and retesting issues. In J.A. Weekley & R.E. Ployhart (Eds.), Situational Judgment Tests: Theory, Measurement and Application (pp. 205 232). Mahwa h, NJ: Lawrence Erlbaum Associates.
68 House, R.J. (1971). A path goal theory of leader effectiveness. Administrative Science Quarterly, 16, 321 328. House, R.J. (1996). Path Goal theory of leadership: Lessons, legacy, and a reformulated theory. Leaders hip Quarterly, 7, 323 352. House, R.J. & Mitchell T.R. (1974). Path Goal theory of leadership. Journal of Contemporary Business, 3, 81 97. House, R.J., Shane, S.A. & Herold, D.M. (1996). Rumors of the death of dispositional research are vastly exaggerat ed. Academy of Management Review, 21, 203 224. International Personality Item Pool: A Scientific Collaboratory for the Development of Advanced Measures of Personality Traits and Other Individual Differences (http://ipip.ori.org/). Internet Web Site. Judge T.A., Piccolo, R.F. & Ilies, R. (2004). The forgotten ones? The validity of consideration and initiating structure in leadership research. Journal of Applied Psychology, 89, 36 51. Kahai, S.S., Sosik, J.J. & Avolio, B.J. (1997). Effects of leadership style and problem structure on work group process and outcomes in an electronic meeting system environment. Personnel Psychology, 50, 121 146. Kolz, A.R., McFarland, L.A., & Silverman, S.B. (1998). Cognitive ability and job experience as predictors of w ork performance. Journal of Psychology: Interdisciplinary & Applied, 132, 539 548.
69 Kristof Brown, A., Barrick, M.R. & Franke, M. (2002). Applicant impression management: Dispositional influences and consequences for recruiter perceptions of fit and simil arity. Journal of Management, 28, 27 46. Lee, S., Choi, S.K., & Choe, I. K. (2005). Issues in measuring situated cognition: Cases of situational judgment tests. Paper presented at the 20 th meeting of the Society for Industrial and Organizational Psycho logy, Dallas, Texas. McCrae, R.R. & Costa, P.T., Jr. (1985). Updating NormanÂ’s Â“adequate taxonomyÂ”: Intelligence and personality dimensions in natural language and in questionnaires. Journal of Personality and Social Psychology, 49, 710 721. McCrae, R. R. & Costa, P.T., Jr. (2003). Personality in Adulthood: A Five Factor Theory Perspective, 2 nd Ed., New York: Guilford Press. McCrae, R.R., Costa, P.T., Jr. & Busch, C.M. (1986). Evaluating comprehensiveness in personality systems: The California Q Set an d the Five Factor Model. Journal of Personality, 54, 430 446. McDaniel, M.A., Morgeson, F.P., Finnegan, E.B., Campion, M.A., & Braverman, E.P. (2001). Use of situational judgment tests to predict job performance: A clarification of the literature. Journ al of Applied Psychology, 86 730 740. McDaniel, M.A., Schmidt, F.L., & Hunter, J.E. (1988). Job experience correlates of job performance. Journal of Applied Psychology, 73, 327 330. McDaniel, M.A., & Nguyen, N.T. (2001). Situational judgment tests: A r eview of practice and constructs assessed. International Journal of Selection and Assessment, 9 103 113.
70 McElreath, J., & Vasilopoulos, N.L. (2002). Situational judgment: What is most or least likely to happen here? Paper presented at the 17 th meeting of the Society of Industrial and Organizational Psychology, Toronto, Ontario, Canada. McFarland, L.A. & Ryan, A.M. (2000). Variance in faking across noncognitive measures. Journal of Applied Psychology, 85, 812 821. Motowidlo, S.J., Dunnette, M.D. & Car ter, G.W. (1990). An alternative selection procedure: The low fidelity simulation. Journal of Applied Psychology, 75, 640 647. Motowidlo, S.J., Hooper, A.C. & Jackson, H.L. (2006). A theoretical basis for situational judgment tests. In J.A. Weekley & R .E. Ployhart (Eds.), Situational Judgment Tests: Theory, Measurement and Application (pp. 57 82). Mahwah, NJ: Lawrence Erlbaum Associates. Motowidlo, S.J. & McDaniel, M.A. (2005). Situational judgment tests part 2: A theory based on implicit trait polic ies. Paper presented at the 20 th meeting of the Society of Industrial and Organizational Psychology, Los Angeles, California. Motowidlo, S.J. & Tippins, N. (1993). Further studies of the low fidelity simulation in the form of a situational inventory. Jo urnal of Occupational & Organizational Psychology, 66, 337 344. Mowry, H.W. (1957). A measure of supervisory quality. Journal of Applied Psychology, 41, 405 408.
71 Norman, W.T. (1963). Toward an adequate taxonomy of personality attributes: Replicated fact or structure in peer nomination personality ratings. Journal of Abnormal and Social Psychology, 66, 574 583. Nunnally, J.C. & Bernstein, I.H. (1994). Psychometric Theory, 3 rd Ed. New York: McGraw Hill. Ones, D.S. & Vishwesvaran, C. (1998). The effects o f social desirability and faking on personality and integrity assessment for personnel selection. Human Performance, 11, 245 269. Oshagbemi, T. (2004). Age influences on the leadership styles and behaviour of managers. Employee Relations, 26, 14 29. Plo yhart, R.E. & Weekley, J.A. (2006). Situational judgment: Some suggestions for future science and practice. In J.A. Weekley & R.E. Ployhart (Eds.), Situational Judgment Tests: Theory, Measurement and Application (pp. 345 350). Mahwah, NJ: Lawrence Erlba um Associates. Quinones, M.A., Ford, J.K. & Teachout, M.S. (1995). The relationship between work experience and job performance: A conceptual and meta analytic review. Personnel Psychology, 48, 887 910. Schmidt, F.L. & Hunter, J.E. (1998). The validit y and utility of selection methods in personnel psychology: Practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262 274.
72 Schmidt, F.L. & Hunter, J.E. (1993). Tacit knowledge, practical intelligence, gener al mental ability and job knowledge. Current Directions in Psychological Science, 2, 8 9. Schriescheim, C.A., Tepper, B.J. & Terault, L.A. (1994). Least preferred coworker score, situational control, and leadership effectiveness: A meta analysis of conti ngency model performance predictions. Journal of Applied Psychology, 79, 561 573. Silvester, J., Anderson Gough, F.M., Anderson, N.R. & Mohamed, A.R. (2002). Locus of control, attributions and impression management in the selection interview. Journal of Occupational and Organizational Psychology, 75, 59 76. Somech, A. (2003). Relationships of participative leadership with relational demography variables: A multi level perspective. Journal of Organizational Behavior, 24, 1003 1018. Sternberg, R.J. (1985 ). Implicit theories of intelligence, creativity and wisdom. Journal of Personality and Social Psychology, 49, 607 627. Sternberg, R.J. & Wagner, R.K. (1993). The g ocentric view of intelligence and job performance is wrong. Current Directions in Psycho logical Science, 2, 1 5. Sternberg, R.J., Wagner, R.K., Williams, W.M., & Horvath, J.A. (1995). Testing common sense. American Psychologist, 50, 912 927. Stogdill, R.M. & Coons, A.E. (1957). Leader Behavior: Its Description and Measurement. Columbus, O H: Ohio State University, Bureau of Business Research.
73 Sturman, M.C. (2003). Searching for the Inverted U Shaped Relationship between time and performance: Meta analyses of the experience /performance, tenure/performance, and age/performance relationships. Journal of Management, 29, 609 640. Styhre, A. (2004). Rethinking knowledge: A Bergsonian critique of the notion of tacit knowledge. British Journal of Management, 15, 177 188. Tett, R.P., Jackson, D.N. & Rothstein, M. (1991). Personality measures as predictors of job performance: A meta analytic review. Personnel Psychology, 44, 703 743. U.S. Department of Labor Bureau of Labor Statistics Standard Occupational Codes ( http://www.bls.gov/oes/current/oes_stru.htm ). Internet Web Site. Vishwesvaran, One s, D.S. & Hough, L.M. (2001). Do impression management scales in personality inventories predict managerial job performance ratings? International Journal of Selection and Assessment, 9, 277 289. Wagner, R.K. (1987). Tacit knowledge in everyday intellig ent behavior. Journal of Personality and Social Psychology, 52, 1236 1247. Wagner, R.K. & Sternberg, R.J. (1991). Tacit Knowledge Inventory for Managers San Antonio, TX: The Psychological Corporation. Weekly, J.A. & Jones, C (1997). Video based Situat ional Testing. Personnel Psychology, 50, 25 49.
74 Weekly, J.A., Ployhart, R.E. & Holtz, B.C. (2006). On the development of situational judgment tests: Issues in item development, scaling and scoring. In J.A. Weekley & R.E. Ployhart (Eds.), Situational J udgment Tests: Theory, Measurement and Application (pp. 157 182). Mahwah, NJ: Lawrence Erlbaum Associates. Yukl, G. (1994). Leadership in Organizations (3 rd Ed.). Englewood Cliffs, NJ: Prentice Hall.
76 Appendix A: Instructional Manipulation I nformation Instructions to Participants ( Participative/Supportive Culture Condition): Please carefully read the following information about the company where you are applying and keep it in mind as you respond to this assessment: ABC Company has a well known reputation for being employee friendly and having strong family values. It is common knowledge that Joseph Meyers, the CEO of ABC Company, worked his way up from the mailroom to make ABC what it is today: a Fortune 500 company with one of the highest employee satisfaction ratings in the business. At ABC, the philosophy is, behind every good manager, there is a team of great people. The leadership style at this company is Participative and Supportive. That means that managers should allow their subordi nates to contribute ideas and even question management, if necessary. It also means that providing personal and professional support to employees is very important. Managers at ABC are expected to be highly supportive of their employees and to provide a ba lance between the demands of work and family. Your employees are encouraged to contribute to the planning and execution of tasks. ABC places a lot of weight on feedback from subordinates when they evaluate performance and award raises. You received the fo llowing confidential email from a friend of a friend (whom youÂ’ve met a few times) who works at ABC Company: Hey, excited you might be working with us at ABC! It is such a great company to work for. The environment is so friendly, and supervisors are alwa ys supportive. You really get the feeling that weÂ’re a team, from the very top to the newest employee. Everyone looks out for everyone else, and everyoneÂ’s voice is heard and acknowledged. Let me know if I can put in a good word for you! Instructions to P articipants ( Directive/Achieving Culture Condition): Please carefully read the following information about the company where you are applying and keep it in mind as you respond to this assessment: XYZ Company has a well known reputation for being aggress ive and competitive. At XYZ, the "bottom line" and profitability always come first. It is common knowledge that Joseph Meyers, the CEO of XYZ Company, doesn't like failure. He has had to lay off a lot of employees and step on a lot of toes to make XYZ Comp any what it is today: a Fortune 500 company with one of the highest profit margins in the business. At XYZ, the philosophy is, managers are accountable; successful managers are well rewarded, and
77 Appendix A: Instructional Manipulation Information (Contin ued) unsuccessful managers are gone. The leadership style at this company is Directive and Achievement oriented. That means that getting results and meeting goals is very important. When your performance evaluations come around, the biggest question is "H ow much did you increase profitability this year?" Managers at XYZ are expected to delegate duties to their employees and make sure they are doing what they are supposed to from day to day, because at the end of the day, what gets done or doesn't get done is management's responsibility. You received the following confidential email from a friend of a friend (whom youÂ’ve met a few times) who works at XYZ Company: Hey, excited you might be working with us at XYZ! It is such a great company to work for. WeÂ’ re not one of those touchy feely companies, but everyone knows what they need to do and we always get the job done. No one wastes a lot of time trying to get consensus. Management knows what theyÂ’re doing and you canÂ’t argue with their results. Let me know if I can put in a good word for you! Instructions to Participants ( Neutral/Control Condition): Please carefully read the following information about the company where you are applying and keep it in mind as you respond to this assessment: NYT is a Fort une 500 company with over ten thousand employees nationwide. Business is growing and NYT will soon expand into international markets. You received the following confidential email from a friend of a friend (whom youÂ’ve met a few times) who works at NYT Co mpany: Hey, excited you might be working with us at NYT! It is such a great company to work for. Let me know if I can put in a good word for you!
78 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions Welcome! This assessment involves th e use of two measures: a "situational judgment test" or SJT, and an opinion survey. An SJT is a kind of test that presents you with questions about how employees should react in realistic work situations. An opinion survey asks you to provide your honest thoughts and experiences. In this case, the opinion survey will ask some questions about what you thought of the SJT as well as some questions about you. What to Expect Imagine that you are applying for a job in management with a real company. As part o f your application, the company has asked you to take the following situational judgment test, the SJT 280. As you take the SJT 280, please respond to the questions as though you really want this management job and it is very important to you that you get it. Remember, everything you will see for the first part of this assessment is based on a real company and a real employment test. So please try your best! To help you do your best on this assessment, you will begin by reading some information about the company where you are applying. During the SJT part of this assessment (26 questions), it will be up to you to read each question carefully and choose One Most Effective Response and One Least Effective Response based on what you know about this compan y.
79 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) Remember to try your best, just as you would if you were really applying for a job! During the opinion survey part of this assessment (31 questions) you will be asked t o answer some questions about yourself and about the SJT assessment that you just finished. If you need to quit at any time, you can always exit and return to this assessment later. When you click on the link to this site, your computer will automatically return you to where you left off (you must be using the same computer). Before we begin the assessment, on what date (day of the month) were you born? 1 st 9 th 10 th 19 th 20 th 31 st (Note: Instructional Manipulation Information is presented here) Choose o ne Most Effective Response and one Least Effective Response based on what you know about this company. 1) You ask an experienced employee to do a particular task. The employee responds curtly, Â“That's not my job.Â” What should you do? Ask the employee i f something is bothering him or her. Do the task yourself but discipline the employee later. Explain why the task is important and ask the employee to reconsider. Get some one else to do it and talk with the employee later. Insist that it is part of the employee's job and see that he/she does it. 2) Of the following, which one method would good managers use most often for monitoring and controlling the work of employees? A ctivity status reports by employees. Feedback from others familiar with the employees' work. Hands on inspection. Impromptu telephone calls and meetings. Time and action c alendars. This instruction is repeated for the first 3 situational judgment items. It is presented in this Appendix only once for the sake of brevity.
80 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) 3) At the end of the day, an employee's car won't start. She is in a hurry to pick up her children. You have not finished closing the office for the evening. Wha t should you do? Allow the employee to use the phone while you finish closing the office. Lend the employee cab fare so she can get her children. Take the employee in your car to get her children, then come back and close the office. 4) Of the following, which do you feel is the most effective way for a Manager to improve communications with Employees? Adopt an "open door" policy. Ask employees a lot of questions. Be very visible and accessible. Schedule regular group meetings. Schedule regular, one on o ne meetings with employees. 5) One of the Managers reporting to you is reluctant to hold his employees accountable for results. He is too willing to accept reasons for why things can't get done according to standard. What should you do? Counsel the Manag er regarding his performance. Hold a meeting with the Manager and his employees regarding the importance of meeting performance expectations. Train the Manager on how to set performance standards and follow up with employees. 6) You are a new Manager. Jus t before you took over, one of the supervisors working under you was promoted into greater responsibility. She is highly intelligent, but not very experienced. You start to get lots of complaints from the employees that she is inflexible and has a philosop hy of "my way or the highway." Employee turnover in her area has been rising since she got there. What should you do? Check it out with a few employees. Consult with the prior Manager. Do an employee survey and review the results with this supervisor. Hav e a small meeting with this supervisor and some of the persons complaining. Tell this supervisor what you've been hearing. 7) You are a Manager. An employee keeps showing up for work late. She is otherwise a good Employee. However, the other employees are noticing that she's coming in late, and it's setting a bad example. Since she started coming in late, you've been trying to find out why. She hasn't been willing to tell you until now: her husband has become a serious problem. She has already been referre d to counseling. What else should you do?
81 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) Ask her what help she needs. Be empathetic but explain the need to be on time. Consult with your boss. Give her more time to work things out. Offer to change her schedule. 8) You are informing the employees in your work unit that they have just won an award for achieving the best quality record in the company. What should you say? "Great job team! Now that we have quality taken car e of, let's strive to achieve the same recognition for quantity." "Great job team! See how far a little hard work and dedication can take you?" "This is a proud moment for all of us and clear evidence that your hard work has been recognized in this company ." "This is a proud moment for all of us. At the same time I know we can achieve even higher quality standards. Let's show them what we can really do." 9) A serious problem has arisen with a project that your work team is currently working on. What should you say? "For some reason we seem to be having a problem concerning this project. Here are my thoughts." "Let's analyze the project step by step and determine what caused the problem." "The good news is that we can solve the problem. Now here's what we a ll need to do." "We seem to be having some problems with our current project. Does anyone have any suggestions?" 10) One of your newer employees is not pulling her weight in the sales department. For the second month in a row, she has not sold the require d amount of goods. What should you do? Ask one of her experienced coworkers to coach her and help her develop better sales skills. Give her a month to improve; if she doesnÂ’t do better by then, talk to her about whether this company is the right fit for h er. Give her some Â“easy customersÂ” that will make her sales numbers look better until she can get in the swing of the job. Ignore the problem; people usually do better after they get some experience. 11) A top performer in your department has just asked y ou for next week off due to a death in the family. Your whole department will be needed next week to prepare an important report for your companyÂ’s CEO, and usually no one is given time off during that week. What should you say?
82 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) Â“IÂ’m sorry I canÂ’t spare you for a whole week. The best I can do is three days. We really need you for this report.Â” Â“Take the week off. But make sure you find someone to cover your portion of the report that is due next week.Â” Â“Take the week off. I wouldnÂ’t do this for just anyone, but you are one of my top performers and it is an emergency situation.Â” 12) Of the following, what should a manager do when introducing a new policy that is likely to b e unpopular? Call a group meeting to introduce and discuss the policy. Meet with each employee individually to discuss the policy. Post the policy on a bulletin board and invite questions. Send a memo to each employee explaining the policy. Sound out empl oyee opinions before announcing the policy. 13) In a staff meeting, you propose that a new project be handled in the usual way, but one of your employees (that you donÂ’t always get along with) interrupts to say he doesnÂ’t think that the Â“usual wayÂ” will w ork in this case. What should you say? Â“Do you have a better idea? If so, you should have mentioned it to me before this meeting.Â” Â“Okay, letÂ’s hear your plan out, and if it sounds good, IÂ’ll expect you to take the lead on this project.Â” Â“Please donÂ’t int errupt. If you have a different idea, letÂ’s talk about it in my office.Â” Â“The usual way has worked for a long time, and IÂ’m going to make an executive decision here to at least give it a try this time.Â” 14) Jill is one of your hardest workers, but she som etimes has trouble getting along with others in the department. She recently came to you with a complaint about Tom, one of your newest employees. Jill reports that Tom is dragging down the rest of the department, taking frequent breaks and generally not g etting much work done. You decide to confront Tom with this information. His immediate reply is, Â“IÂ’ll bet Jill was the one who came to you about this. SheÂ’s been giving me a hard time ever since I started here.Â” What should you say? Â“IÂ’ve spoken to sever al people in the department, and they are all giving me the same story: youÂ’re not getting your work done.Â” Â“It doesnÂ’t matter who it was. IÂ’m concerned to hear this kind of information about you, Tom.Â” Â“Matter of fact, it was Jill. YouÂ’d do better to be m ore like her, and complain less about her, Tom.Â” Â“Tom, it seems like you havenÂ’t been happy here since you started. What can I do to help you get established with us?Â” Â“Why donÂ’t you tell me your side of the story, Tom.Â”
83 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) 15) Betsy has three children, and it seems like they are always getting sick. The result is that Betsy has to take a lot of time off on short notice to be with her children. It seems her work is suffering b ecause of it, too. Some of your other employees are getting upset about BetsyÂ’s absences, though no one has complained to you directly yet. What should you do? Let Betsy know that her time off is almost used up, and that taking additional time off could r esult in a written warning, leading up to termination. Let Betsy know that her work is suffering and that if her work doesnÂ’t improve, you will have no choice but to give her a written warning, leading up to termination. Let Betsy know you have set a meeti ng to discuss some possible changes she could pursue to improve her commitment to work, such as using onsite child care, getting a babysitter, or getting her significant other to watch the kids when they are sick. Let Betsy know you have set a meeting to d iscuss some possible scheduling options that will better fit her lifestyle, such as working part time, working a 4 day week, or changing to evening shifts. Wait until someone approaches you about the problem. If no one has made an official complaint, it pr obably isnÂ’t serious enough to worry about. 16) Mike is probably one of the most productive people in your department. He is nearly always punctual, organized and diligent, and the work he does is first rate. He has a reputation in the department for bein g sort of a loner and not being too talkative with other employees. The other day in the break room, you see Mike sitting by himself drinking coffee. What should you do? Go over and strike up a conversation with Mike about an interesting movie you saw rec ently. Go over and strike up a conversation with Mike about how things are going at work. Go over and tell Mike, Â“I wish I had 10 other employees just like you.Â” Say, Â“Hi Mike,Â” as you pass him on your way out. Talk to one or two of your employees about he lping Mike feel more included in the company. 17) Pat, one of your employees, just shouted at a very important client over the telephone, and everyone in the office heard it. You ask whatÂ’s going on, and Pat says, Â“IÂ’d rather not talk about it.Â” What shou ld you say? Â“Okay, but I canÂ’t tolerate you talking like that in this office.Â” Â“Okay, but I canÂ’t tolerate you talking like that to one of our most important clients.Â” Â“Okay, but IÂ’m going to have to ask you to call back the client and apologize right now .Â” Â“Okay, but IÂ’m going to have to ask you to leave the area until you settle down.Â” Â“Okay, but if you tell me whatÂ’s going on, maybe I can help.Â”
84 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) 18) The home office in formed you that you must have a workplace seminar next week to comply with company regulations. They will send a speaker on one of the following topics, all of which have been well received by employees in the past. You must make a decision right away so t hey can schedule the speaker. What is your choice? A Stitch in Time: Maximizing Workplace Efficiency Balancing the Scales: How to Find a Happy Medium Between Work and Family Brainstorming: Getting the Most Out of EveryoneÂ’s Ideas Closing the Deal: Selling Yourself and Your Company 19) An employee who is supervised by one of your subordinates has asked to talk with you. He says your subordinate is guilty of some possible violations of company policy. This is the first you have heard of any problem with the subordinate in question. In the company, employees are expected to take all concerns to their immediate supervisor before going to anyone else. The employee who wants to talk with you has not discussed this matter with his supervisor (your subordinate) be cause of its sensitive nature. What should you do? Refuse to meet with this employee until he has first discussed this matter with his supervisor (your subordinate). Meet with this employee, but only with your subordinate present. Meet with this employee to discuss the matter, and then decide whether you need to meet with your subordinate. Meet with your subordinate to discuss the matter, and then decide whether you need to meet with this employee. 20) You are a manager, and you have an outstanding work t eam. Lately, you have been getting complaints from your team. They say they seem to get assigned every project that comes along. You feel that this is probably true. What should you do? Talk with your director and ask him if he thinks your team is getting more than their fair share of new assignments. Talk with your directorÂ’s supervisor and tell him that your team is getting more assignments than any other team in the organization. Talk with other managers in your department. Explain that new projects sho uld be divided up evenly. Tell your team that because they are the best in the department, sometimes they have to pick up the slack for other work teams.
85 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) 21) You are a newly assigned manager. Your department director has given you guidelines on how to run your work team. What input should you ask from your team? Ask your team what they think is the best way to handle projects, but keep in mind that it is your decision to make. Tell your team exactly how you want them to handle projects. Do not ask for their input unless a problem comes up. Assuming the team has been performing well let them decide how they would like to be run. Try to merge what they say with your dire ctorÂ’s guidelines. Determine how the team has been performing, combine that information with your directorÂ’s guidelines, then run the team in the manner that you believe is best. 22) One of your employees, Bob, has always performed his work in an excellen t manner, and seemed to be happy working for you. Today he came to you and said he wanted to transfer to a different work team and if possible to a different department. What should you do? Talk with Bob. If you find that he really wants to transfer, then help him to do so. Agree to begin the transfer if thatÂ’s what Bob really wants. Do not process the transfer too quickly, and in the meantime, try to help Bob with whatever is leading him to request the transfer. Let Bob know very clearly how pleased you h ave been with his performance and how much youÂ’d like him to remain on your team. Then ask him if there are any problems you might be able to help with. Tell Bob you would like him to hold off on the transfer for 1 month to see if together you can solve wh atever problems are behind his request for transfer. 23) As a manager, you have noticed that one of your employees, Joan, has been taking a lot of her own time to help out another employee. What should you do? Commend Joan and recommend her for special r ecognition. Tell Joan that you appreciate her help, but also mention that you expect her not to neglect her own duties. Talk with Joan to find out if the other employee has some special problem that you should know about. Let Joan know you appreciate her h elp. Tell Joan that you appreciate her help, but that the other employee needs to learn to do his/her own work. 24) One of your employees is performing an assigned task exactly the way it is supposed to be done, but he is taking entirely too much time to get it done. Your work team has several more tasks to do today, and you told the employee to work faster so that the entire group will not have to stay late. An hour later, when you came back to see what he had accomplished, you found that he had not don a nything since the last time you talked with him. Because of this, you and your whole team will have to work late tonight to meet a deadline. What should you do?
86 Appendix B: Situational Judgment Inventory (SJT 280) and Instructions (Continued) Talk to the employee and find out if anything is bothering him. Assign another employee to assist him until the job is completed. Meet with him later to find out why he was so slow. Put this employee on a different task and have one of your better employees finish up this task. Tell the whole team that everyone is staying late because of one employee. Hopefully, this will create peer pressure and speed up his performance. 25) You are a manager. Yesterday, your work team finished up a difficult and exhausting 2 week p roject. This morning, your department director overlooked several other teams and assigned an important Â“rushÂ” project to your team. Your employees have given you a lot of negative feedback about this assignment. What do you do? Advise your director that your employees have just finished a difficult project and you do not feel they are capable of giving their best work on a new Â“rushÂ” project at this time. Persuade your director to select another team for the new project. Determine why your team was select ed (because of superior performance or just necessity), then inform your team of the reason for their selection and the importance of going forward with the new project. Ask your director to consider giving your team a different project since they just fin ished a difficult project and you do not feel they are capable of giving their best work on this new "rush" project. If your director won't reconsider, take the matter to your director's supervisor. 26) You have a new director who has been the head of you r department for about 3 months. He just told you that he has recommended one of your employees for a promotion. You told him that you have worked with this employee for over a year and that you donÂ’t think the employee is ready for a promotion. Your direc tor says he has already made up his mind, but you are sure that the employee is not ready. What should you do? Present your new director with all the documentation you have to support your views of the employee in question. Ask other managers in your depa rtment what they have done in similar circumstances. Present your documentation in an informational email to your directorÂ’s supervisor.
87 Appendix C: Secondary Inventory (SPE 30) and Instructions Thank you for taking the time to complete the SJT 280. The next part of this research will ask you some questions about the instructions you read, the situational judgment test (SJT 280) you just completed, and some additional questions about you and your personality. At this point, you DO NOT need to imagine you are trying to get a job. Answer the following questionnaire about yourself and your true opinions as honestly as you can. Choose the response that best describes how much you agree or disagree with each statement about how you responded to the SJT. [The following anchors are presented above each set of items: 1 = I Very Much Disagree 2 = I Somewhat Disagree 3 = I am Neutral 4 = I Somewhat Agree 5 = I Very Much Agree] When answering this inventory, I found myself focusing on situations I have been in be fore, even though they did not happen at work. 1 2 3 4 5 When answering this inventory, I chose the answers that I did because I have a pretty good idea of what will happen as a result of each choice. 1 2 3 4 5 When answering this inventory, I chose the answers that I did because they seemed like the Â“right thing to do.Â” 1 2 3 4 5 When answering this inventory, I did not answer the way I normally would, because I was asked to answer as though I really wanted the job. 1 2 3 4 5
88 Appendix C: Se condary Inventory (SPE 30) and Instructions (Continued) Choose the response that best describes how much you agree or disagree with each statement about your everyday life I base most of my decisions in everyday life on how I have made decisions in the past. 1 2 3 4 5 I think I have a good idea of what a managerÂ’s job involves. 1 2 3 4 5 I think I am the kind of person who would enjoy a full time job in management. 1 2 3 4 5 Most of my work experience has been with a company that is very similar to th e company in the description that I read. 1 2 3 4 5 I expect that the values at most companies are similar to the values at the company in the description that I read. 1 2 3 4 5 The description of the imaginary company provided some clues about how I sho uld answer the situational judgment items in order to get the job. 1 2 3 4 5 The email from the imaginary coworker provided some clues about how I should answer the situational judgment items in order to get the job. 1 2 3 4 5 The company that I work for now is very similar to the company in the description that I read. 1 2 3 4 5 In my experience, most managers would fit right in at the company in the description that I read. 1 2 3 4 5 Most of my experience as a manager has been in a company very sim ilar to the description that I read. (if you have no management experience, please answer this question as an employee). 1 2 3 4 5 Choose the response that best describes how much you agree or disagree with each statement about your everyday life I fee l little concern for others. 1 2 3 4 5 I am interested in people. 1 2 3 4 5 I never insult people. 1 2 3 4 5 I sympathize with others' feelings. 1 2 3 4 5 I have a soft heart. 1 2 3 4 5 I am not really interested in others. 1 2 3 4 5 I take time out for others. 1 2 3 4 5 I feel others' emotions. 1 2 3 4 5 I make people feel at ease. 1 2 3 4 5 I am not interested in other people's problems. 1 2 3 4 5 I tend to trust other people, for the most part. 1 2 3 4 5 I will listen to others if I believe th ey can make a contribution. 1 2 3 4 5 I am the kind of person whom others can trust. 1 2 3 4 5 I seldom have all the answers. 1 2 3 4 5
89 Appendix C: Secondary Inventory (SPE 30) and Instructions (Continued) Choose the answer that best describes your w ork experience: I have worked part time or full time in a management job (with any company) for: 0 years (never) less than 1 year total 1 to 3 years total 3 to 5 years total more than 5 years total Choose the answer that best describes your experienc e with your current company: I have been with my current company (in any job) for: 0 years (I am not currently employed) less than 1 year 1 to 3 years 3 to 5 years more than 5 years If you are currently employed, please select the industry that best describes your current job, or select "Other" and describe your current job in the box provided. If you are not currently employed, please choose NOT EMPLOYED. NOT EMPLOYED Communications Energy Finance Government Healthcare Manufacturing Retail Social Services Science/Technology Transportation Other (Please specify Below):
90 Appendix D: Manipulation Pilot Inventory For each statement below, indicate whether you 1 = Strongly Disagree 2 = Moderately Disagree 3 = Neither Disagree nor Agree 4 = Moderately Agree 5 = Strongly Agree SD MD N MA SA After reading the description of this company, it sounds like a place where employees are told what is expected of them. 1 2 3 4 5 After reading the description of this company, it sounds like a place where emplo yees are part of the decision making process. 1 2 3 4 5 After reading the description of this company, it sounds like a place where employees are told how to perform their jobs. 1 2 3 4 5 After reading the description of this company, it sounds like a pl ace where employees are asked for suggestions. 1 2 3 4 5 After reading the description of this company, it sounds like a place where leadership is friendly and approachable. 1 2 3 4 5 After reading the description of this company, it sounds like a place where leaders show concern for their employeesÂ’ well being. 1 2 3 4 5 After reading the description of this company, it sounds like a place where employees are expected to perform at their highest level. 1 2 3 4 5 After reading the description of this co mpany, it sounds like a place where leaders set challenging goals for their employees. 1 2 3 4 5 After reading the description of this company, it sounds like the kind of place I would like to work. 1 2 3 4 5
91 Appendix E: Situational Judgment Inventory Pilot Rating Instructions Used to Create Conditional Response Keys and Measure Reliability Imagine that you work as a manager at a company that is creating a test to help select new managers. Your company has asked you to help create a key for this new t est. A key is a set of correct or best responses to a test against which a candidateÂ’s responses can be compared. First, some information will be presented to help you understand how things work at your company. This should help you understand how to c reate a key that will be specific to your company. Then, it will be up to you to read each question on the test and rate the response choices for each question using the following scale: A Very Effective Response A Somewhat Effective Response A Somewhat Ineffective Response A Very Ineffective Response You will find that the questions on the test have 3, 4, or 5 response choices for you to rate. Therefore, you might not use every rating for each question, or you might use some ratings more than once f or each question. Please try to rate every response choice for every question. Try to find at least one effective and one ineffective response for each question. Finally, do not over think any response choice or question; just go with your first impre ssion using your best judgment. Thank you for your participation!
92 Appendix F: Significant SJT Item level Chi Square and Phi Values Most Effective Item Least Effective 2 p Phi R 2 2 p Phi R 2 39.17 0.000 0.315 0.10 1 28.5 0.000 0.268 0.07 32.63 0.000 0.387 0.15 2 3 11.44 0.022 0.17 0.03 4 5 6 7 16.44 0.036 0.204 0.04 8 28.7 0.000 0.269 0.07 9 13.67 0.034 0.186 0.03 10 15.55 0.016 0.198 0.04 11 12 13 22.49 0.004 0.238 0.06 14 16.75 0.033 0.206 0.04 15 18.78 0.016 0.218 0.05 16 35.27 0.000 0.298 0.09 17 16.45 0.036 0.204 0.04 22.51 0.001 0.238 0.06 18 28.15 0.000 0.267 0. 07 19 15.42 0.017 0.197 0.04 27.91 0.000 0.265 0.07 20 15.44 0.017 0.197 0.04 14.43 0.025 0.191 0.04 21 22 23 17.55 0.007 0.211 0.04 24 18.9 0.004 0.218 0.05 10.47 0.033 0.163 0.03 25 26 12 Number S ignificant 10 46.15% Percent Significant 38.46%
93 Appendix G: Item Level Frequency Tables for Managers and Students MANAGERS (N = 258) Most Effective Least Effective Response # Participative Directive Control Participative Directive Control N=74 N=88 N=96 N=74 N=88 N=96 1a 25 .7% 9.1% 14.6% 6.8% 17.0% 10.4% b 0.0% 0.0% 0.0% 43.2% 52.3% 49.0% c 66.2% 61.4% 72.9% 1.4% 0.0% 1.0% d 4.1% 4.5% 1.0% 18.9% 14.8% 21.9% e 4.1% 25.0% 11.5% 29.7% 15.9% 17.7% 2a 18.9% 17.0% 10.4% 6.8% 12.5% 8.3% b 5.4% 0.0% 1.0% 41.9% 38.6% 53.1% c 74.3% 73.9% 82.3% 4.1% 2.3% 0.0% d 0.0% 0.0% 0.0% 35.1% 33.0% 31.3% e 1.4% 9.1% 6.3% 12.2% 13.6% 7.3% 3a 77.0% 90.9% 92.7% 14.9% 4.5% 2.1% b 9.5% 2.3% 5.2% 14.9% 34.1% 21.9% c 13.5% 6.8% 2.1% 70.3% 61.4% 76.0% 4a 1 0.8% 13.6% 11.5% 10.8% 13.6% 7.3% b 0.0% 1.1% 0.0% 64.9% 63.6% 62.5% c 50.0% 40.9% 35.4% 4.1% 1.1% 2.1% d 1.4% 2.3% 1.0% 20.3% 20.5% 28.1% e 37.8% 42.0% 52.1% 0.0% 1.1% 0.0% 5a 9.5% 14.8% 4.2% 32.4% 33.0% 37.5% b 0.0% 6.8% 7.3% 64.9% 60.2% 59.4% c 9 0.5% 78.4% 88.5% 2.7% 6.8% 3.1% 6a 5.4% 4.5% 2.1% 29.7% 19.3% 22.9% b 14.9% 11.4% 7.3% 14.9% 19.3% 20.8% c 44.6% 34.1% 34.4% 9.5% 11.4% 10.4% d 17.6% 22.7% 19.8% 28.4% 27.3% 28.1% e 17.6% 27.3% 36.5% 17.6% 22.7% 17.7% 7a 29.7% 18.2% 18.8% 2.7% 2.3% 7 .3% b 50.0% 73.9% 72.9% 4.1% 2.3% 4.2% c 1.4% 2.3% 0.0% 21.6% 18.2% 11.5% d 0.0% 0.0% 0.0% 56.8% 62.5% 68.8% e 20.3% 5.7% 8.3% 14.9% 14.8% 8.3% 8a 9.5% 17.0% 9.4% 43.2% 33.0% 39.6% b 2.7% 4.5% 7.3% 33.8% 36.4% 39.6% c 62.2% 44.3% 47.9% 6.8% 6.8% 7.3 % d 25.7% 34.1% 35.4% 16.2% 23.9% 13.5% 9a 1.4% 9.1% 0.0% 40.5% 44.3% 45.8% b 27.0% 48.9% 38.5% 16.2% 19.3% 10.4% c 14.9% 18.2% 19.8% 32.4% 14.8% 31.3% d 56.8% 23.9% 41.7% 10.8% 21.6% 12.5%
94 Appendix G: Item Level Frequency Tables (Continued) 10a 86. 5% 79.5% 89.6% 5.4% 2.3% 0.0% b 10.8% 18.2% 9.4% 12.2% 2.3% 4.2% c 2.7% 2.3% 1.0% 12.2% 17.0% 9.4% d 0.0% 0.0% 0.0% 70.3% 78.4% 86.5% 11a 28.4% 38.6% 31.3% 29.7% 29.5% 38.5% b 55.4% 45.5% 57.3% 13.5% 12.5% 9.4% c 16.2% 15.9% 11.5% 56.8% 58.0% 52.1% 12a 71.6% 61.4% 60.4% 1.4% 3.4% 1.0% b 25.7% 33.0% 36.5% 4.1% 2.3% 2.1% c 0.0% 0.0% 0.0% 58.1% 42.0% 47.9% d 0.0% 4.5% 2.1% 10.8% 23.9% 15.6% e 2.7% 1.1% 1.0% 25.7% 28.4% 33.3% 13a 1.4% 4.5% 3.1% 41.9% 40.9% 53.1% b 79.7% 71.6% 80.2% 4.1% 4.5% 0.0% c 5.4% 11.4% 9.4% 37.8% 38.6% 31.3% d 13.5% 12.5% 7.3% 16.2% 15.9% 15.6% 14a 2.7% 1.1% 0.0% 10.8% 6.8% 3.1% b 14.9% 21.6% 26.0% 6.8% 3.4% 1.0% c 0.0% 4.5% 0.0% 81.1% 84.1% 87.5% d 40.5% 20.5% 24.0% 1.4% 4.5% 6.3% e 41.9% 52.3% 50.0% 0.0% 1.1% 2.1% 1 5a 10.8% 12.5% 9.4% 9.5% 6.8% 3.1% b 2.7% 10.2% 10.4% 6.8% 2.3% 0.0% c 33.8% 50.0% 41.7% 1.4% 1.1% 3.1% d 52.7% 26.1% 38.5% 1.4% 4.5% 0.0% e 0.0% 1.1% 0.0% 81.1% 85.2% 93.8% 16a 20.3% 5.7% 11.5% 9.5% 11.4% 7.3% b 51.4% 70.5% 69.8% 0.0% 1.1% 0.0% c 4 .1% 8.0% 1.0% 24.3% 21.6% 16.7% d 5.4% 8.0% 5.2% 41.9% 35.2% 54.2% e 18.9% 8.0% 12.5% 24.3% 30.7% 21.9% 17a 6.8% 10.2% 5.2% 23.0% 18.2% 18.8% b 5.4% 12.5% 1.0% 6.8% 12.5% 21.9% c 4.1% 8.0% 3.1% 39.2% 35.2% 34.4% d 8.1% 5.7% 10.4% 28.4% 21.6% 21.9% e 75.7% 63.6% 80.2% 2.7% 12.5% 3.1% 18a 21.6% 38.6% 27.1% 32.4% 12.5% 16.7% b 25.7% 13.6% 11.5% 23.0% 56.8% 41.7% c 27.0% 9.1% 20.8% 21.6% 19.3% 24.0% d 27.0% 38.6% 40.6% 23.0% 11.4% 17.7%
95 Appendix G: Item Level Frequency Tables (Continued) 19a 5.4% 3.4% 2.1% 63.5% 76.1% 79.2% b 6.8% 9.1% 6.3% 27.0% 9.1% 10.4% c 81.1% 73.9% 80.2% 2.7% 5.7% 2.1% d 6.8% 13.6% 11.5% 6.8% 9.1% 8.3% 20a 74.3% 48.9% 55.2% 2.7% 11.4% 4.2% b 1.4% 0.0% 0.0% 43.2% 44.3% 54.2% c 12.2% 27.3% 26.0% 12.2% 9.1% 7.3% d 12.2% 23.9% 18.8% 41.9% 35.2% 34.4% 21a 21.6% 13.6% 13.5% 5.4% 9.1% 6.3% b 1.4% 8.0% 1.0% 86.5% 78.4% 86.5% c 31.1% 12.5% 21.9% 6.8% 10.2% 6.3% d 45.9% 65.9% 63.5% 1.4% 2.3% 1.0% 22a 20.3% 18.2% 24.0% 13.5% 13.6% 8.3% b 1.4% 2.3% 1.0% 50.0% 48.9% 54.2% c 77.0% 76.1% 72.9% 0.0% 2.3% 0.0% d 1.4% 3.4% 2.1% 36.5% 35.2% 37.5% 23a 14.9% 8.0% 9.4% 4.1% 12.5% 10.4% b 6.8% 20.5% 16.7% 36.5% 25.0% 25.0% c 74.3% 62.5% 68.8% 8.1% 11.4% 6.3% d 4.1% 9.1% 5.2% 51.4% 51.1% 58.3% 24a 33.8% 28.4% 28.1% 1.4% 5.7% 0.0% b 60.8% 62.5% 70.8% 1.4% 1.1% 0.0% c 4.1% 3.4% 1.0% 4.1% 3.4% 3.1% d 1.4% 5.7% 0.0% 93.2% 89.8% 96.9% 25a 24.3% 12.5% 9.4% 9.5% 22.7% 16.7% b 74.3% 87.5% 89.6% 8.1% 2.3% 1.0% c 1.4% 0.0% 1.0% 82.4% 75.0% 82.3% 26a 97.3% 94.3% 95.8% 0.0% 3.4% 1.0% b 1.4% 2.3% 3.1% 27.0% 28.4% 21.9% c 1.4% 3.4% 1.0% 73.0% 68.2% 77.1% = Summed Six Keyed Responses STUDENTS (N = 138) Most Effective Least Effective Response # Participative Directive Control Participative Directive Control N=40 N=53 N=45 N=40 N=53 N=45 1a 15.0% 9.4% 25.0% 15.0% 43.4% 18.2% b 0.0% 1.9% 0.0% 40.0% 43.4% 38.6% c 65.0% 37.7% 47.7% 2.5% 1.9% 0.0% d 10.0% 3.8% 6.8% 10.0% 3.8% 15.9% e 10.0% 47.2% 22.7% 32.5% 7.5% 29.5%
96 Appendix G: Item Level Frequency Tables (Continued) 2a 17.5% 26.4% 13.6% 15.0% 17.0% 25.0% b 20.0% 1.9% 6.8% 35.0% 39.6% 36.4% c 52.5% 50.9% 59.1% 7.5% 0.0% 2.3% d 5.0% 1.9% 13.6% 30.0% 20.8% 25.0% e 5.0% 18.9% 9.1% 12.5% 22.6% 13.6% 3a 52.5% 54.7% 52.3% 15.0% 11.3% 18.2% b 30.0% 30.2% 29.5% 12.5% 17 .0% 18.2% c 17.5% 15.1% 20.5% 72.5% 71.7% 65.9% 4a 27.5% 20.8% 18.2% 12.5% 9.4% 9.1% b 0.0% 1.9% 2.3% 72.5% 64.2% 56.8% c 32.5% 34.0% 22.7% 0.0% 5.7% 18.2% d 10.0% 17.0% 13.6% 10.0% 17.0% 13.6% e 30.0% 26.4% 45.5% 5.0% 3.8% 4.5% 5a 7.5% 9.4% 20.5% 6 2.5% 64.2% 50.0% b 32.5% 35.8% 36.4% 27.5% 22.6% 36.4% c 60.0% 54.7% 45.5% 10.0% 13.2% 15.9% 6a 10.0% 1.9% 4.5% 35.0% 28.3% 22.7% b 7.5% 5.7% 15.9% 20.0% 17.0% 25.0% c 42.5% 45.3% 43.2% 2.5% 3.8% 6.8% d 20.0% 32.1% 27.3% 12.5% 18.9% 18.2% e 20.0% 15 .1% 11.4% 30.0% 32.1% 29.5% 7a 12.5% 9.4% 18.2% 10.0% 9.4% 18.2% b 42.5% 50.9% 34.1% 15.0% 15.1% 13.6% c 5.0% 1.9% 6.8% 37.5% 26.4% 15.9% d 2.5% 5.7% 2.3% 30.0% 37.7% 50.0% e 37.5% 32.1% 40.9% 7.5% 11.3% 4.5% 8a 7.5% 0.0% 11.4% 30.0% 41.5% 38.6% b 7 .5% 13.2% 2.3% 32.5% 34.0% 31.8% c 57.5% 49.1% 56.8% 7.5% 9.4% 9.1% d 27.5% 37.7% 31.8% 30.0% 15.1% 22.7% 9a 7.5% 3.8% 9.1% 42.5% 35.8% 52.3% b 32.5% 54.7% 34.1% 12.5% 5.7% 13.6% c 10.0% 18.9% 13.6% 32.5% 34.0% 31.8% d 50.0% 22.6% 45.5% 12.5% 26.4% 4 .5% 10a 87.5% 69.8% 75.0% 0.0% 0.0% 0.0% b 12.5% 28.3% 20.5% 7.5% 1.9% 6.8% c 0.0% 1.9% 6.8% 10.0% 18.9% 22.7% d 0.0% 0.0% 0.0% 82.5% 79.2% 72.7% 11a 25.0% 30.2% 34.1% 47.5% 32.1% 38.6% b 45.0% 54.7% 56.8% 12.5% 15.1% 4.5% c 30.0% 15.1% 11.4% 40.0% 52.8% 59.1%
97 Appendix G: Item Level Frequency Tables (Continued) 12a 67.5% 69.8% 70.5% 2.5% 1.9% 4.5% b 17.5% 18.9% 18.2% 10.0% 5.7% 9.1% c 7.5% 1.9% 2.3% 40.0% 34.0% 38.6% d 2.5% 3.8% 4.5% 32.5% 22.6% 34.1% e 5.0% 5.7% 6.8% 15.0% 35.8% 15.9% 13a 2 .5% 5.7% 11.4% 47.5% 47.2% 59.1% b 67.5% 52.8% 65.9% 5.0% 18.9% 6.8% c 17.5% 24.5% 20.5% 32.5% 22.6% 20.5% d 12.5% 17.0% 4.5% 15.0% 11.3% 15.9% 14a 2.5% 7.5% 4.5% 7.5% 3.8% 11.4% b 20.0% 35.8% 29.5% 2.5% 1.9% 4.5% c 0.0% 1.9% 2.3% 82.5% 83.0% 77.3% d 42.5% 22.6% 22.7% 5.0% 3.8% 6.8% e 35.0% 32.1% 43.2% 2.5% 7.5% 2.3% 15a 5.0% 7.5% 9.1% 10.0% 9.4% 15.9% b 5.0% 26.4% 15.9% 2.5% 9.4% 2.3% c 50.0% 32.1% 40.9% 0.0% 0.0% 4.5% d 37.5% 34.0% 36.4% 0.0% 3.8% 6.8% e 2.5% 0.0% 0.0% 87.5% 77.4% 72.7% 16a 30.0% 22.6% 27.3% 2.5% 5.7% 4.5% b 40.0% 47.2% 43.2% 5.0% 1.9% 6.8% c 7.5% 13.2% 15.9% 20.0% 18.9% 13.6% d 7.5% 9.4% 2.3% 47.5% 37.7% 50.0% e 17.5% 7.5% 13.6% 25.0% 35.8% 27.3% 17a 2.5% 15.1% 9.1% 22.5% 20.8% 20.5% b 2.5% 15.1% 2.3% 12.5% 9.4% 6.8% c 17.5% 32.1% 22.7% 27.5% 24.5% 31.8% d 0.0% 5.7% 6.8% 32.5% 26.4% 29.5% e 77.5% 32.1% 61.4% 5.0% 18.9% 13.6% 18a 15.0% 34.0% 36.4% 12.5% 11.3% 18.2% b 22.5% 15.1% 27.3% 27.5% 50.9% 27.3% c 55.0% 32.1% 31.8% 5.0% 5.7% 11.4% d 7.5% 18.9% 6.8% 55.0% 32 .1% 45.5% 19a 12.5% 7.5% 4.5% 62.5% 56.6% 61.4% b 20.0% 20.8% 27.3% 20.0% 11.3% 25.0% c 45.0% 52.8% 56.8% 10.0% 18.9% 9.1% d 22.5% 18.9% 13.6% 7.5% 13.2% 6.8% 20a 50.0% 43.4% 27.3% 2.5% 17.0% 11.4% b 22.5% 7.5% 9.1% 20.0% 18.9% 25.0% c 25.0% 28.3% 5 6.8% 7.5% 13.2% 6.8% d 2.5% 20.8% 9.1% 70.0% 50.9% 59.1%
98 Appendix G: Item Level Frequency Tables (Continued) 21a 25.0% 26.4% 22.7% 2.5% 7.5% 9.1% b 5.0% 7.5% 4.5% 87.5% 71.7% 72.7% c 17.5% 17.0% 20.5% 7.5% 17.0% 11.4% d 52.5% 49.1% 54.5% 2.5% 3.8% 9.1% 22a 7.5% 15.1% 9.1% 12.5% 20.8% 29.5% b 22.5% 9.4% 9.1% 42.5% 37.7% 31.8% c 57.5% 66.0% 79.5% 7.5% 5.7% 0.0% d 12.5% 9.4% 4.5% 37.5% 35.8% 40.9% 23a 12.5% 1.9% 13.6% 45.0% 32.1% 22.7% b 25.0% 34.0% 18.2% 10.0% 18.9% 20.5% c 52.5% 54.7% 56.8% 7. 5% 5.7% 6.8% d 10.0% 9.4% 13.6% 37.5% 43.4% 52.3% 24a 27.5% 9.4% 22.7% 5.0% 20.8% 6.8% b 57.5% 64.2% 68.2% 0.0% 5.7% 2.3% c 10.0% 15.1% 11.4% 10.0% 7.5% 2.3% d 5.0% 11.3% 0.0% 85.0% 66.0% 90.9% 25a 40.0% 43.4% 25.0% 17.5% 18.9% 38.6% b 45.0% 45.3% 5 4.5% 27.5% 30.2% 31.8% c 15.0% 11.3% 22.7% 55.0% 50.9% 31.8% 26a 72.5% 77.4% 77.3% 7.5% 3.8% 4.5% b 17.5% 9.4% 13.6% 32.5% 49.1% 50.0% c 10.0% 13.2% 11.4% 60.0% 47.2% 47.7% = Summed Six Keyed Responses
99 Appendix H: Student SJT Item Level Chi Sq uare and Phi Values Student Most Effective Item Student Least Effective 2 p Phi R 2 2 p Phi R 2 21.98 0.005 0.399 0.159 1 21.815 0.005 0.398 0.158 20.686 0.008 0.387 0.150 2 8.501 0.386 0.248 0.062 0.414 0.981 0.055 0.003 3 1.392 0.846 0.1 0.010 6.448 0.597 0.216 0.047 4 11.14 0.194 0.284 0.081 4.375 0.358 0.178 0.032 5 3.055 0.549 0.149 0.022 8.166 0.417 0.243 0.059 6 3.494 0.9 0.159 0.025 5.849 0.664 0.206 0.042 7 9.029 0.34 0.256 0.066 10.293 0.113 0.273 0.075 8 3.328 0.767 0.155 0.024 11.448 0.075 0.288 0.083 9 10.036 0.123 0.27 0.073 7.296 0.121 0.23 0.053 10 4. 064 0.397 0.172 0.030 5.665 0.226 0.203 0.041 11 5.491 0.241 0.199 0.040 2.72 0.951 0.14 0.020 12 8.577 0.379 0.249 0.062 7.359 0.289 0.231 0.053 13 7.635 0.266 0.235 0.055 8.765 0.362 0.252 0.064 14 5.023 0.755 0.191 0.036 11.143 0.194 0.284 0.081 15 11.612 0.169 0.29 0.084 6.916 0.546 0.224 0.050 16 4.443 0.815 0.179 0.032 24.851 0.002 0.429 0.184 17 4.871 0.771 0.188 0.035 13.416 0.037 0.312 0.097 18 10.184 0.117 0.272 0.074 3.774 0.707 0.165 0.027 19 6.145 0.407 0.211 0.045 21.862 0.001 0.398 0.158 20 7.535 0.274 0.234 0.055 0.875 0.99 0.08 0.006 21 6.162 0.405 0.211 0.045 8.268 0.219 0.245 0.060 22 6.659 0.354 0.22 0.048 7.492 0.278 0.233 0.054 23 5.918 0.432 0.207 0.043 10.655 0.1 0.278 0.077 24 12.874 0.045 0.305 0.093 4.903 0.297 0.188 0.035 25 7.971 0.093 0.24 0.058 1.429 0.839 0.102 0.010 26 3.378 0.497 0.156 0.024 5 Number Significant (bold) 2 19.2% Percent Significant 7.7%
100 Appendix I: Manager SJT Item Level Chi Square and Phi Values Mgr Most Item Mgr Least 2 p Phi R 2 2 p Phi R 2 23.49 0.001 0.302 0.091 1 11.21 0.19 0.208 0.043 13.742 0.033 0.231 0.053 2 9.62 0.293 0.193 0.037 13.01 0.011 0.225 0.051 3 19.124 0.001 0.272 0.074 6.96 0.541 0.164 0.027 4 6.816 0.557 0.163 0.027 11.642 0.02 0.212 0.045 5 2 .693 0.61 0.102 0.010 11.011 0.201 0.207 0.043 6 3.646 0.888 0.119 0.014 16.88 0.01 0.256 0.066 7 9.65 0.29 0.193 0.037 8.776 0.187 0.184 0.034 8 4.474 0.613 0.132 0.017 28.772 0.0001 0.334 0.112 9 13.017 0.043 0.225 0.051 4.335 0.363 0.13 0.017 10 16 .264 0.012 0.251 0.063 3.669 0.453 0.119 0.014 11 2.475 0.649 0.098 0.010 6.997 0.321 0.165 0.027 12 9.437 0.307 0.191 0.036 5.317 0.504 0.144 0.021 13 6.867 0.333 0.167 0.028 20.374 0.009 0.281 0.079 14 11.816 0.16 0.214 0.046 16.024 0.042 0.249 0.06 2 15 16.942 0.031 0.256 0.066 19.641 0.012 0.276 0.076 16 9.324 0.316 0.19 0.036 16.843 0.032 0.256 0.066 17 17.357 0.027 0.259 0.067 18.774 0.005 0.27 0.073 18 23.598 0.001 0.302 0.091 4.023 0.674 0.125 0.016 19 14.377 0.026 0.236 0.056 15.132 0.019 0.242 0.059 20 9.05 0.171 0.187 0.035 19.49 0.003 0.275 0.076 21 2.893 0.822 0.106 0.011 2.109 0.909 0.09 0.008 22 5.636 0.465 0.148 0.022 9.958 0.126 0.196 0.038 23 7.741 0.258 0.173 0.030 9.759 0.135 0.194 0.038 24 8.36 0.213 0.18 0.032 9.172 0.057 0.189 0.036 25 11.186 0.025 0.208 0.043 2.116 0.714 0.091 0.008 26 4.619 0.329 0.134 0.018 13 Number Significant 8 50.0% Percent Significant 30.8%
About the Author Jonathan Adam Shoemaker has held an interest in genera l psychology since childhood. He completed his B.A. with Honors in Psychology at The College of William & Mary in 1995. He completed his M.S. in Applied Psychology at Georgia College & State University in 1997. He has taught numerous courses as a gradu ate student and provided selection, training and development consulting services for various organizations. He continues to be interested in selection, training, organizational culture, and occupational health psychology.