USF Libraries
USF Digital Collections

Educational policy analysis archives

MISSING IMAGE

Material Information

Title:
Educational policy analysis archives
Physical Description:
Serial
Language:
English
Creator:
Arizona State University
University of South Florida
Publisher:
Arizona State University
University of South Florida.
Place of Publication:
Tempe, Ariz
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Education -- Research -- Periodicals   ( lcsh )
Genre:
non-fiction   ( marcgt )
serial   ( sobekcm )

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - E11-00348
usfldc handle - e11.348
System ID:
SFS0024511:00348


This item is only available as the following downloads:


Full Text
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 a22 u 4500
controlfield tag 008 c20039999azu 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E11-00348
0 245
Educational policy analysis archives.
n Vol. 11, no. 50 (December 27, 2003).
260
Tempe, Ariz. :
b Arizona State University ;
Tampa, Fla. :
University of South Florida.
c December 27, 2003
505
Exploration of the pay levels needed to attract students with mathematics, science and technology skills to a career in K-12 teaching / Anthony Milanowski.
650
Education
x Research
v Periodicals.
2 710
Arizona State University.
University of South Florida.
1 773
t Education Policy Analysis Archives (EPAA)
4 856
u http://digital.lib.usf.edu/?e11.348


xml version 1.0 encoding UTF-8 standalone no
mods:mods xmlns:mods http:www.loc.govmodsv3 xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govmodsv3mods-3-1.xsd
mods:relatedItem type host
mods:identifier issn 1068-2341mods:part
mods:detail volume mods:number 11issue 50series Year mods:caption 20032003Month December12Day 2727mods:originInfo mods:dateIssued iso8601 2003-12-27



PAGE 1

1 of 25 A peer-reviewed scholarly journal Editor: Gene V Glass College of Education Arizona State University Copyright is retained by the first or sole author, who grants right of first publication to the EDUCATION POLICY ANALYSIS ARCHIVES EPAA is a project of the Education Policy Studies Laboratory. Articles appearing in EPAA are abstracted in the Current Index to Journals in Education by the ERIC Clearinghouse on Assessment and Evaluation and are permanently archived in Resources in Education Volume 11 Number 50December 27, 2003ISSN 1068-2341An Exploration of the Pay Levels Needed to Attract Students with Mathematics, Science and Technology Skills to a Car eer in K-12 Teaching Anthony Milanowski University of Wisconsin-MadisonCitation: Milanowski, A. (2003, December 27). An ex ploration of the pay levels needed to attract stude nts with mathematics, science and technology skills to a car eer in K-12 teaching, Education Policy Analysis Archives, 11 (50). Retrieved [Date] from http://epaa.asu.edu/epa a/v11n50/.Abstract In an exploratory study (Note 1) of the role of sal ary level and other factors in motivating undergraduate math, sci ence, and technology majors to consider a career as a K-12 te acher, the salary level students said would motivate them to c onsider a career in teaching was related to the salary expect ed in their chosen non-teaching occupation, but not to three of the Big 5 personality dimensions of extroversion, agreeablene ss, and openness, nor concern for others or career risk ave rsion. An annual starting salary 45% above the local average would attract 48% of the sophomore students and 37% of the junior s. Focus group results suggested that low pay was an importa nt reason for not considering K-12 teaching, but that perceived j ob demands and abilities and interests were also important rea sons for not being attracted to a teaching career. In a number of states and districts around the coun try, concerns have been raised about an actual or potential shortage of teachers. Though the often-cited figure of 2 million new teachers needed gives an erroneous picture of a gen eral teacher shortage (Felter, 1997; National Association of State Boards of Education,1 998), there are appear to be real local

PAGE 2

2 of 25 shortages caused by reforms such as class size redu ctions (Curran et al, 2000), shortages in poor urban or rural districts (Hirsh, 2001), and shortages in specific subject areas, notably math, science, and special education (Hirsh 2001; Recruiting New Teachers, 2000; American Federation of Teachers, 1998, Pullen 1999). For an individual district, given enough resources, shortages can probably be overcome though the use of standard human resources manageme nt techniques such as broader and more intensive recruitment efforts and increasi ng compensation (directly through wage and benefit increases or indirectly though lig hter teaching loads or better working conditions). But from a state or national perspecti ve, in order to increase the total supply of teachers, especially in shortage subjects like math or science, broader policy interventions may be needed. A common policy prescription is rais ing teacher compensation, both to attract more new entrants to the teacher labor mark et and to retain teachers (Eastin, 2000, Hare et al, 2001; National Commission on Mathematic s and Science Teaching for the 21stCentury, 2000). There is evidence that increased co mpensation does increase the size of the teacher pool. (Murnane et al, 1991; Darling-Ham mond et al, 1999; Ferris and Winkler, 1988). But there appears to be little current resea rch on the size of the increase in beginning salaries that would be needed to draw a s ignificantly larger number of entrants to one shortage area, math and science teaching.A casual comparison of beginning salaries for K-12 teaching to those for occupations that now attract college graduates with the knowledge to teach math and science suggests that beginning salaries for teaching would have to rise substantially to draw workers away from these occupations. A review of average salary offer s to new graduates reported by the National Association of Colleges and Employers (200 2) shows that median annual salary offers for majors in mathematics, accounting, scien ce (except biological and life science) engineering, or computer science were the low 40’s to low 50’s. Median offers to elementary and secondary education majors reported by the same source were in the $29-30 thousand per year range (see Appendix Table 1). The American Federation of Teachers (2001) salary survey reported an average t eacher beginning salary of $27,989 for 1999-2000, and beginning salaries in math and s cience related occupations ranging from $37,688 for accounting to $47,112 for engineer ing. These numbers suggest that substantial increases in beginning salaries would b e needed to make teaching competitive for graduates with math or science knowledge, even taking into account the shorter work year. However, these comparisons may overstate the size of the increase in entry salaries needed. It is not necessary that all of those who m ight otherwise choose one of these careers be attracted to K-12 teaching, only enough to remedy the shortage. But how much is enough?The salary level needed to attract enough qualified people to eliminate the shortage of math and science teachers depends on the elasticity of the supply of potential new teachers with respect to entry salary. This depends in turn on factors such as the number of people with the necessary skills and abilities, the interests and values of these people, the availability of alternative career opportunitie s, and the entry salaries in those opportunities. Given enough people with the needed knowledge and skills, personality, interest, and value factors are likely to be import ant in the success of a policy of increasing beginning salaries, because occupational choice is not just a matter of comparing beginning salaries. The vocational guidance literat ure (e.g. Holland, 1985, Dawis, 1990) tends to portray occupational choice as heavily inf luenced by personality, values, and interests. One implication of this is that increasi ng teaching entry salaries to the level of other occupations may not substantially increase th e supply of new math and science teachers because people who have chosen other occup ations have done so based on their interests, personality, and values, which are relatively stable and formed early in life. There is some evidence that teachers and those aspi ring to be teachers have a different

PAGE 3

3 of 25 pattern of values or personality from those pursuin g other careers. For example, Ben-Shem and Avi-Itzah (1991) found that Freshmen e ntering helping professions (including education) tended to be more compassiona te, caring, and empathetic than those entering business. Bradley (1983) found that British pre-university students who had chosen teaching put more importance on the opportun ity to work with children and the social value of the job while those choosing other occupations indicated that salary and non-routine work were more important. Tusin (1999) reviewed studies of motivations for entering teaching, and concluded that the desire to work with children and provide service to society were important attractors, along with wo rking in a subject of interest and the school time schedule. Teaching may be less attracti ve to those without these interests and values.We can think of the pool of people who have the ski lls but who are not interested in teaching at the current pay levels as containing tw o stylized types. People of the first type have personalities, interests and values somewhat l ike those of people who have chosen teaching, but prefer higher salaries. These people would accept a teaching job if the pay were closer to or the same as what is available in alternative math, science, or technology occupations. The second type includes those who hav e personalities, interests, and values that incline them fairly strongly to non-teaching c areers in science, math, and technology. This group would likely require a compensating diff erential to attract them to teaching, a salary rate above the current entry levels for their chosen occupati ons, other things equal. The policy prescription of raising teacher entry sa laries to the level of competing math, science, or technology occupations is more likely t o be successful, at lower cost, if there are a substantial number of Type 1 people with the required knowledge and skills and who would like to take a teaching job, but who are dete rred primarily by low salaries. But if values, personality traits, and interests largely d etermine who will choose teaching, and these factors are set relatively early in people’s lives, the size of this group is limited, and may be quite small.As well as influencing the salary level at which a person with the needed skills would consider K-12 math or science teaching, personality interest, and value factors may also be relevant to the recruitment of potential K-12 ma th and science teachers. It may be worthwhile to target people with values, interests, and personalities like those of teachers in recruitment messages, as well as to communicate higher pay. Current attempts to encourage students to try teaching seem to emphasiz e altruism. One example is Teach for America, though the benefits to the student are als o emphasized. Also, it is possible that people whose values interests and personalities are more like those who have chosen to teach may be more likely to be good teachers and st ay in teaching. The research described here explored the potential influence of higher entry salaries and personality and value factors on the attractiveness of K-12 teaching to undergraduate college students interested in majors that involve math and science. The specific research questions of interest were: Can math, science, or technology majors be attracte d to a career in K-12 teaching by higher starting salaries, given the potential impor tance of interests, values, and personality traits in occupational choice? 1. How much of an increase in starting salary would mo tivate a substantial proportion of math, science, or technology majors to consider a c areer in K-12 teaching? 2. Do personality characteristics and work values infl uence the salary level that would motivate math, science or technology majors to cons ider a career in K-12 teaching? 3. Do other characteristics of the teaching job beside s salary level reduce its 4.

PAGE 4

4 of 25 attractiveness to math, science, and technology maj ors? These questions were addressed first by holding a s eries of focus group discussions with university students expected to have the knowledge to teach math or science, and then by asking such students to respond to a survey on whic h they were presented with a set of pay levels at which they might find K-12 teaching a ttractive, as well as personality and value items. Focus GroupsMethodFocus groups were held to provide background inform ation and explore several issues about the attractiveness of teaching careers to stu dents. The groups were held in the first part of the Spring semester. Each session was sched uled for 90 minutes, but some concluded in less time. One set of four groups was held with students in math, science, applied science, or engineering majors, while anoth er four groups were held with students interested in a teaching career, in order to see if there were any differences in reasons for occupational choice between the two groups.Participants. Participants were freshmen and sophomores at a lar ge Midwestern research university. Freshmen and sophomores were invited to participate because we wanted the opinions of students who had not yet made a strong commitment to a major, and who could more easily switch to an Education major. At the junior and senior level, switching to an Education major often requires lengthening the t ime to graduate in order to take required education courses. Most recruitment was do ne in large math courses required for math, science, and technology majors. With the inst ructors’ permission, a brief presentation was made to the class explaining the b asic requirements of participation and the incentive offered ($25). Occupations represente d in the math, science, and technology focus groups included actuarial science ,astronomy, chemistry, biochemistry, computer science, engineering (mostly computer engineering), medicine, nursing, pharmacy, and veterinary science. In all 19 students participated Three groups had five or six participants, and one had three, due to three stude nts not showing up for that session. Of the participating students, 13 were female, six wer e male. The average age was 18.3 years and the average self-reported GPA was 3.26.As a contrast, focus groups were also conducted wit h a group of education and pre-education majors at the same university. Freshm an and sophomores who were interested in a teaching career were recruited both from math classes and via an email sent to their university email address. Invitations explained the purpose of the research, the basic requirements of participation, and the in centive offered ($25). Email addresses were obtained from the school of education’s studen t advising department. In all, 23 student participated. All groups had between five a nd seven participants. Nineteen participants were female, four were male. The avera ge age was 19.7 years and the average self-reported GPA was 3.4.Focus Group Procedure and Analysis. Prior to beginning the discussions, participants completed a short questionnaire covering demographi c information and expected salaries. Discussions were then conducted by a professional f acilitator, who asked the students to respond to a set of discussion questions. The discu ssion protocol began with questions about the occupations the students were interested in pursuing and the reasons for their interest, then introduced questions about preferenc es for pay for performance systems (for another project), and ended with questions about te aching as a career and the influence of salary level on the attractiveness of a teaching ca reer. The sessions were taped and the

PAGE 5

5 of 25 tapes transcribed. Transcripts were content-analyze d by the author to identify and record broad themes in the discussion of the questions.Results As expected, students interested in mathematics, sc ience, and technology majors had many and varied reasons for occupational choice. Th e reasons fit into four broad categories: interest in the content of the occupati on (more common with those interested in pure science majors), expectations that the student would like content of the job or be good at it, extrinsic rewards such as pay (more com mon with those interested in engineering and pharmacy majors), and helping or wo rking with people (more common in the health majors). While altruism was not a primar y theme of most participants’ comments, several mentioned their chosen occupation s as making a contribution to society.Participants were also asked about their perception s of the job of a K-12 teacher, and if they had considered K-12 teaching as a career. In e very group, the theme of low pay was represented in most participants’ comments. Low pay was the reason most often stated for not considering a teaching career, a reason that wa s stated immediately by many participants. However, in each group, some particip ants also characterized the job as requiring skills or attributes they did not have, s uch as patience, enjoyment of working with people, and ability to connect with students. Other s mentioned concerns like discomfort being responsible for others, student behavior prob lems, taking work home, frustration with kids who don’t “get it”, and the monotony of teachi ng the same material year after year. While some participants mentioned the opportunity t o continue to learn, others appeared to feel that teaching at the K-12 level was an inte llectual dead end. An interesting theme, expressed mostly by computer-related majors, was a lack of up-to-date equipment in schools.Many participants commented on the importance of te aching to society and to students. Most seemed to respect their teachers and teaching as a career choice. Some participants expressed the attitude that they could do better th an teaching, and that teachers were somewhat unfortunate to be stuck in that career. A few participants interested in pure science careers indicated that teaching at the K-12 level was a fall-back career for them, and several others said they would consider teachin g, but only at the post-secondary level. Teaching at the university levels was perceived as more attractive than at the K-12 level. Interestingly, none of the participants indicated a n interest in teaching for a short time before they began their “real” careers, but several indicated they might consider it after they had done well in their current career choice.The major reasons given for their occupational choi ce by the prospective teachers during their focus groups were enjoyment of children or wo rking with them, the desire to influence or help children, past success at tutoring or coach ing, and a schedule that would accommodate family demands and provide summers off. Many also cited their own teachers as models. Prospective teachers generally recognized that their occupation was not that highly paid. (The median expected first ye ar annual salary was $25,000, quite close to the statewide average starting salary in t he area.) Many commented, however, that the benefits (e.g. health insurance, pension, and time-off) and job security were good. Many of the female participants mentioned the conve nience of the K-12 teaching schedule for those raising families. Interestingly, many als o explicitly stated that teachers should not choose the field for money. Statements expressing t he idea that it would be wrong to go into teaching for money were common. One participan t went so far as to say that salaries should be kept low to keep people who would be moti vated to choose teaching by money out of the profession.

PAGE 6

6 of 25 Students interested in math, science, and technolog y majors were also asked if they would seriously consider K-12 teaching as a career if it paid 20% more than they expected to be paid for a job in the occupation in which they were currently most interested. Ten of the 19 participants said they would. Of those who indicate d they would not, there were two dominant reasons: concerns about job requirements o f teaching (e.g. working with children, need for patience) and commitment to the current occupational choice. Many giving the latter reason emphasized that their chos en career was something they had their hearts set on doing. When participants were asked i f they would consider K-12 teaching as a career if it paid 50% more than they expected to be paid for a job in their currently-chosen occupation, 13 of the 19 participa nts said they would. The remaining 6 continued to say either that they did not have the talent or were committed to their current choice.Discussion The focus group results suggest both that increasin g the pay level of K-12 teaching has the potential for increasing the supply of people w ith math and science knowledge who would be willing to consider teaching, and that the re is a limit to how effective higher salaries might be in attracting new entrants to thi s occupation. That the majority of participants identified low pay as a reason they wo uld not consider a career in K-12 teaching, and that most would consider a K-12 teach ing career if it paid substantially more than their current choice, is consistent with the c ommon wisdom that higher pay would increase supply. That most participants would consi der teaching if salaries were higher than they expected to receive in their current occu pation of choice suggests that interests, values, and personality factors do get traded-off w ith salary. These results suggest that K-12 teaching could bid away labor from math, scien ce and technology occupations, given a sufficient pay level or compensating differential However, the fact that a substantial minority of the participants said they would not co nsider a career change even to make substantially higher salaries, either because they did not have the needed skills or abilities or because they are committed to another career cho ice, suggests that even large salary level increases have limitations in expanding the p ool of potential K-12 math and science teachers. This latter finding is consistent with th e literature that emphasizes the influences of interests, values and personality on career choi ce. Education and pre-education majors, as expected, expressed more altruistic reasons for their career choice and recognized that they were not preparing for a highly-paid career. T hey appeared to have chosen teaching in spite of its salary level. This is what one woul d expect to find given the current salary levels for teaching.Limitations Four focus groups of 19 people are too small a samp le to do more than suggest how math, science and technology majors might respond to high er teaching salary levels. Also, the salary levels themselves (20% to 50% higher than th ese students expected to receive in their chosen occupation) are unrealistic as policy options. It also should be recognized that the recruitment for the focus groups probably resul ted in participants who were more sure of their future career paths. Because we asked for participants who were interested in math, science and technology majors, we did not get as many of the ‘undecided’ students common in freshman and sophomore classes. There may be many students with interests and abilities in math and science that are not as c ommitted to non-teaching careers, and would be attracted to teaching by salary levels bet ween the current starting level for teaching and the current starting level for math, s cience and technology jobs.

PAGE 7

7 of 25 SurveyAlthough the focus group results suggest that much higher starting salaries would attract math, science, and technology majors to K-12 teachi ng, the effect of incrementally higher salaries is more likely to be of interest to policy makers. Therefore, based in part on the focus group results, another study was designed to explore the effect of more realistic higher starting salary levels on the willingness of students in math, science and technology majors to consider a career in K-12 teaching. The s tudy was also designed to collect data on other factors likely to influence the level of s alary that these students would find attractive. The most obvious factor is the starting pay level these student expect to earn in jobs within the occupation they have currently chos en. Based on the focus group results, it was also expected that altruism, or concern for oth ers might attract students to teaching, and that students who were concerned about working with people or uncomfortable in front of a classroom would be less attracted. Grade point average might also be important, given the Murname et al (1991) finding that better students were more sensitive to financial incentives in making career choices. Gender, too ma y play a role, since women expecting to be primary childcare-givers may be more likely t o appreciate the time schedule of a K-12 teaching job.MethodParticipants. The sample consisted of sophomores and juniors at a large Midwestern public research university. The choice of universit y was based on convenience in this exploratory study. The university registrar’s offic e provided a random sample of 1000 juniors and 1000 sophomores with declared pre-major s (sophomores) or majors (juniors) in mathematics, health occupations (e.g. medicine, pharmacy, nursing, medical technology), engineering, the sciences (e.g. chemis try, biochemistry, physics, biology, zoology), and applied science (e.g. bacteriology, c omputer science). Majors were chosen which required substantial mathematics and/or scien ce coursework, based on the expectation that students in these majors would hav e the interest and aptitude to acquire the content knowledge needed to teach mathematics a nd science. From these samples, 358 sophomores and 300 juniors were selected at ran dom to receive invitations to participate. The number of invitations sent was det ermined by the funds available to provide an incentive to participate. Because studen ts might have changed majors since they declared majors or pre-majors in the Fall, the y were asked to list their current major. Respondents who indicated they were no longer major ing in the areas involving math or science knowledge were excluded from the sample. On e hundred two sophomores and 106 juniors provided enough data to be included in the analyses, or resulting in a response rate of 28% for the sophomore and 35% for the junio rs. The achieved sophomore sample was 60% female, with an average age of 19.3 years. The achieved junior sample was 38% female, with an average age of 20.9 years. Appendix Table 2 shows the self-reported majors of both groups, and the salaries they expect to receive. It should be noted that there were more engineering and fewer science major s in the junior sample, and that the median expected pay was higher in the junior sample Procedure. Students in the sample were invited to participate in a study of job choice, job characteristics, and pay in their major field via a n email message sent to their university email address. They were asked to log on to a web s ite and complete a survey, which would take about 30 minutes. They were also told th at they would receive $15 for completing the survey. Five days after the initial invitation, students who had not responded received a second invitation. A third and final invitation was sent one week after the second. Sophomores were invited to participate in the Fall, then Juniors in the Spring.

PAGE 8

8 of 25 The survey included 159 mostly closed-ended items ( including items relevant to another study). Students responded to the items by clicking on the “radio button” associated with the chosen response option. Students were asked to enter the name of the occupation or job they were planning to get when they finished th eir education. Many of the survey questions were then asked with reference to this oc cupation. Measures. Attractive salary. Students were told that the current average beginn ing salary for K-12 teachers in the state was $26,500 for a 10 mon th work year. (This figure was obtained from the state association of school board s, which maintains a data base of teacher pay schedules of state public school distri cts.) They were also told that this equated to a 12 month salary of $31,800. They were then asked to choose the beginning salary that that would make them seriousl y consider a career in K-12 teaching. Eleven response categories were provided, the lowest labeled $26,500 and the highest $46,500 or more, with intervening categ ories labeled in $2,000 increments from the lowest. Expected salary Students were asked to enter the annual salary th ey expected to be paid in their first job in their chosen occupation. They were asked to enter their best guess even if they were not sure what salary to exp ect. Personality and work-values measures. To explore the potential impact of personality and interest factors on the salary level at which m ath, science, and technology majors would find teaching attractive, three of the ‘Big 5’ personality dimensions were measured: extroversion, agreeableness, and openness (For a discussion of the Big 5, see Mount and Barrick,1995 and Hogan, Hogan, and Roberts, 1996). These dimensions were chosen because according to researc h by Tokar and Swanson (1995), they can serve as proxies for Holland’s voc ational personality typology (Holland, 1985), a commonly-used vocational interes t measure. According to the Holland typology, teachers are characterized as soc ial (the primary type) with elements of the artistic and enterprising types. Ac cording to Tokar and Swanson’s results, the three Big 5 dimensions chosen correlat e with these Holland types and can distinguish among Holland types. In addition, e xtroversion was also expected to be related to the degree to which study participant s would be comfortable working with students and leading classes. One would expect participants who are less extroverted to require a higher salary to consider a career in K-12 teaching. Saucier’s (1994) ‘mini-markers’ of the Big 5 were used. Respo ndents were required to choose how well each of 40 adjectives (e.g. bold, kind, sh y ) applies to them. Each personality dimension was represented by 8 adjectiv es. The response scale ranged from 1 (does not apply to me) to 9 (applies a great deal). Ratings of related adjectives are summed to form the scale for each tr ait. Work values were measured by the concern for others subscale of the Comparati ve Emphasis Scale, designed to measure four workplace values: achievement, concern for others, fairness, and honesty (Meglino and Ravlin, 1998; Ravlin and Megli no,1997). This measure has a forced choice format in which respondents are asked to choose which of two statements described the value they felt should rec eive the greater emphasis if a choice of action were called for. It produces a ran k order of the 4 values, or an interval scale value for one dimension when only on e value dimension is of interest. In this study, the scale representing the value of concern for others was used to represent altruism, and it was expected that greate r concern for others would make it more likely that a student would consider teaching at a lower salary level. In addition, risk aversion in career choice was measured. An 8 i tem scale developed by Judge and colleagues (Judge et al, 1999; Cable and Judge, 1994) was used. Participants were asked to agree or disagree, on a 5-point Liker t scale, with items such as “I am not willing to take risks when choosing a job or co mpany to work for” and “I prefer a

PAGE 9

9 of 25 high security job with a steady salary over one off ering high risks and high rewards”. Risk aversion was included because of the interest in job security expressed by some science and technology majors in the focus gro ups, and the mention of good job security in the education major focus groups.Participants were also asked to rate the importance of 13 factors or conditions that might effect a decision to pursue a teaching career assuming teaching paid as much as they would expect to receive in the occupation t hey were considering. Among the conditions or factors were the extra work it would take to get a teaching license, the prestige of teaching, problems with student behavio r, and the availability of jobs in a desirable location. Based on the focus group result s, items were also included referring to whether the respondent would like work ing with children, whether s/he would be good at teaching, and whether up-to-date e quipment or technology is available is schools. The importance of these condi tions in making a decision was rated on a 1 (not important) to 5 (very important) scale. The intent of including these items was to try to assess the importance of some o f the non-salary factors that might go into a decision to choose a teaching caree r. Other Variables Several other measures were included in the surve y as potential controls for statistical analyses. Grade point aver age was measured by having students report their current GPA by choosing one o f 6 categories ranging from below 2.0 to 3.7-4.0. Respondents were also asked t o indicate their gender, age and whether they planned to go to graduate school befor e beginning their first job in their chosen occupation. Treatment of Missing Values. Though participants were asked to complete all ite ms, sporadic missing values occurred for several of the questions of this somewhat long survey. Missing values in the personality and value measures were imputed using the method of adjusted mean substitution described by R aaijmakers (1999). Data were imputed for 3 cases in the junior group and 5 cases in the sophomore group. In one case in the sophomore group, the expected salary was imp uted based on the median for the major.ResultsPanel 1 of Table 1 shows the number of participants who chose each salary level when asked to choose which would make them seriously con sider a career in K-12 teaching. It also attempts to represent a sort of supply curve b y showing the cumulative percent of each group that would be willing to consider teachi ng at each higher salary level. (This assumes that those who would consider teaching at o ne salary level would also consider it at the higher levels.) Figure 1 plots the cumulativ e percent against the attractive salary for the combined group of participants.Table 1 Number and Cumulative Percent of Participants Who W ould Consider a Teaching Career at Various Teaching Salary LevelsI. All Participants Attractive Annual Salary Level No. of Sophomores Cumulative Percent of Sophs. No. of Juniors Cumulative Percent of Juniors 26,5001211.8%43.8%28,500112.7%14.7%

PAGE 10

10 of 25 30,500416.7%610.4%32,500 (+23%)723.5%817.9%34,500629.4%522.6%36,500938.2%1032.1%38,500 (+45%)1048.0%637.7%40,5001966.7%2056.6%42,500470.6%.561.3%44,500575.5%868.9%46,500+25100.0%33100.0%Total:102106 II. Engineering & Science/Applied Science Participa nts Attractive Annual Salary Level Engineering Soph. & Jr. Cumulative Percent Science & Applied Sci. Soph. & Jr. Cumulative Percent 26,50023.3%1011.4%28,50003.3%213.6%30,50026.7%418.2%32,500 (+23%)413.3%625.0%34,500420.0%328.4%36,500426.7%938.6%38,500 (+45%)433.3%847.7%40,5001456.7%1160.2%42,500361.7%.565.9%44,5005875.0%470.5%46,500+15100.0%26100.0%Total:6088

PAGE 11

11 of 25 Figure 1. Cumulative Percent of Respondents Who Wou ld Consider a Teaching Career at Various Salary Levels (Sophomore and Juni or Respondents Combined). These data give some idea of the increase in starti ng salaries that would be needed to attract a substantial number students majoring in m ath, science, or technology to K-12 teaching.A 23% increase in starting salary would be needed t o attract about 23% of the sophomore participants and about 18% of the juniors. A 45% in crease would be needed to attract 48% of the sophomore participants and about 37% of the juniors. Looking at Figure 1, if one can consider the range of annual salaries from $28, 500 to $38,500 as representing a plausible range of salaries a state or the nation m ight be likely to fund (increases of 7.5 to 45%), there is a fairly linear relationship between salary increases and the increase in the percent of participants who would consider teaching Each $1,000 salary increase is associated with an increase of 3.4 percentage point s in the proportion of respondents willing to consider a career in K-12 teaching. (Not e that if plotted separately, the sophomore curve would be lower and the junior curve higher than that of the two groups combined, but the curves are very close to parallel in the $28,500 to $38,500 range.) Panel 2 of Table 1 shows the number of participants who chose each salary level as attractive for the two largest groups in the sample engineering students and pure and applied science students. A 23% increase in startin g salary would attract about 13% of the engineering student participants and about 25% of t he pure and applied science participants. A 45% increase would attract about 33 % of the engineering students and about 48% of the pure and applied science students. In the range of annual salaries from $28,500 to $38,500, each $1,000 salary increase is associated with an increase of 3.1 percentage points in the proportion of respondents willing to consider a career in K-12 teaching for the engineering students and 3.4% for the pure and applied science students. The salary respondents expected to be paid in their first job in their chosen non-teaching occupation was expected to influence the salary tha t would attract them to teaching. Figures 2 and 3 plot the natural log of the attract ive salary against the natural log of the expected salary for the sophomore and junior partic ipants, respectively. The vertical line

PAGE 12

12 of 25 parallel to the Y axis is set at the value of the t eaching minimum salary level. The “fitted” line is the lowess smoothing of attractive on expec ted salary. These figures show a relationship between expected salary and attractive salary that is not as simple as was expected. Figure 2. Relationship Between the Natural Logarith ms of Expected Beginning Salary and the Teaching Beginning Salary Respondent s Chose as Attractive – Sophomore Respondents Figure 3. Relationship Between the Natural Logarith ms of Expected Beginning Salary and the Teaching Beginning Salary Respondent s Chose as Attractive – Junior RespondentsThe figures show that expected pay level was not as strong a predictor of the attractive pay

PAGE 13

13 of 25 level as might be expected. The correlations betwee n the natural log of the attractive and expected beginning salaries were .41 for the sophom ores and .21 for the juniors. The figures show that at most levels of expected salary there was a wide range of attractive teaching salaries. While in the middle range of bot h variables the relationship between has the expected upward slope (Note 2), there are two i nteresting anomalous groups of participants. In the upper left corners, one group of participants had low expected salaries but claimed to be attracted to teaching only by muc h higher teaching salaries (higher than their expected salaries). In the lower center, anot her group had relatively high expected salaries (compared to teaching) but chose the curre nt average teacher minimum as the salary that would attract them to teaching.Drawing on the focus group results, the first group may represent students who would require a large premium or compensating differentia l to consider teaching because they do not have the skills or personality traits they perc eived to be required, because they regarded some of the job characteristics of teachin g as disagreeable, or because they placed a high value on their current career choice. The second group, those with expected salaries above the teaching starting salary but who would be willing to consider teaching at its current salary level, are somewhat harder to sp eculate about. Some may be students who would consider teaching as a back-up career pla n, or be highly altruistic. One would, however, have expected the altruists to have alread y chosen teaching. Some may have thought teaching paid even less than the minimum th ey were given, and would consider teaching at the level given in the survey. Of cours e, some students in this group could simply have given erroneous or unconsidered respons es. These results support distinguishing between two st ylized ‘types’ of participant: those who would be willing to consider teaching at a lower sa lary than they expected to receive in their current choice, and those willing to consider teaching only for a higher salary than expected. The first type might be thought of as tho se willing to trade some of the salary difference between their current choice and teachin g for a shorter work year, a work schedule more convenient for raising children, or t he opportunity to work in a job satisfying needs for altruism. The second type includes those requiring a compensating differential to teach at the K-12 level. Of the sophomore participa nts, 41% reported an attractive salary for teaching that was above their expected salary. For the juniors, the corresponding percentage was 31%.In order to explore the roles of personality and wo rk values, and the effect of gender and grade point average on the salary level participant s chose as required to get them to consider K-12 teaching, Tobit models were estimated The 2 limit Tobit model was used because of the censoring of high and low values of the dependant variable, due to the use of fixed response options beginning at $26,500 and ending at $46,500. It was expected that participants with higher scores on concern for others, higher scores on extroversion, and higher scores for career risk aversion would ch oose lower attractive salary levels. In addition, participants who were female and had lowe r grade point averages were also expected to choose lower attractive salary levels. Gender (female =1) and intent to go to graduate school before stating the career (intent = 1) were represented by dummy variables. The reported GPA was represented by the numeric value of the response category encompassing a student’s GPA. One outlier was removed from the junior sample. The results are shown in Table 2. (Means, standard deviations, and correlations are shown in Appendix Tables 3 and 4.)Table 2 Tobit Prediction of Attractive Salary for Sophomore and Junior Samples

PAGE 14

14 of 25 Sophomore SampleJunior Sample Independent Variable CoefficientStandard ErrorCoefficientStandard Error Natural Log, Expected Salary.23805*.05798.20801*.07 389 Female-.04763 .05214.04694.05131Intend to Attend Grad School.00119.05158.05500.0468 4 Grade Point Average.01472.01342.00878.01292Career Risk Aversion-.05187.05171-.03231.04558Extroversion.00037.00238.00045.00205Openness.00137.00309-.00215.00287Agreeableness-.00299.00377-.00281.00302Concern for Others-.02361*.01154.00232.01038Chi-square/prob. (9 d.f.) 31.10/.00015.43/0.080 Statistically significant at .05 level or beyond.Expected salary was a significant predictor of the attractive salary in both samples. Concern for others is negatively related to attract ive salary, as expected, in the sophomore sample, but the coefficient is close to 0 and not s ignificant in the junior sample. None of the other predictors had statistically significant coefficients. The extroversion measure, though its coefficient had the expected sign, was b arely larger than its standard error in the sophomore sample and smaller in the junior sample. The model fit the data better for the sophomore sample. Additional analyses using an orde red logit approach showed the same results in terms of the pattern of significant coef ficients, with higher chi-square and pseudo-R-squared values for the sophomore than the junior sample. As mentioned above, participants were also presente d with a list of 13 factors or conditions that might effect a decision to pursue a teaching career, and asked to rate the importance of each factor on a 1 (not important) to 5 (very important) scale, assuming teaching paid as much as they would expect to recei ve in the occupation they were considering. Table 3 shows the means and standard d eviations of the ratings of these factors for each sample.Table 3 Importance of Factors Other Than Pay in Deciding to Pursue a Career in TeachingSophomoresJuniors Reason Mean Standard Deviation Mean Standard Deviation Whether would like working with children 4.570.764.520.75 Whether would be good at teaching4.410.844.420.77

PAGE 15

15 of 25 Need to give up current career plan4.111.144.181.10Ability to continue to work in discipline3.801.083. 961.00 How easy to find job in teaching3.481.003.641.02Jobs available in desired location3.581.173.721.13Level of benefits3.780.943.860.83Opportunities for advancement3.381.193.281.08Problems with student behavior3.141.133.071.14Availability of up-to-date equipment or technology 3.111.072.981.00 Extra work needed to get license2.871.152.651.16Status or prestige of teaching2.431.222.371.13Approval of parents or important others2.401.132.13 1.14 These results tend to be consistent with some of th e focus group results. The high ratings for concerns about being good at teaching and enjoy ing working with children, giving up the current career choice and being able to continu e in the discipline were expected given that these themes were mentioned in the focus group s. Prestige and parental approval were rated as less important (perhaps reflecting so cial desirability). These respondents appear to be more concerned about the aspects of te aching that relate to their interests and abilities than with some of the ‘extrinsic’ fea tures such as benefits and prestige. These results do not, of course, suggest that other facto rs like ease of finding a job or level of benefits are not important, but only that in these samples respondents less concerned about prestige are more likely to be interested in a teaching career at a lower beginning salary level.DiscussionThis study attempted to explore the questions of wh ether undergraduate students with knowledge and interests in math, science, or techno logy could be attracted to a career in K-12 teaching by higher starting salaries, what sal ary levels might be needed to get more of these students into K-12 teaching, whether perso nality and work values would influence the salary level that would attract these students to teaching, and what other characteristics of the teaching job reduce its attr activeness to these students. The findings suggest that students with knowledge a nd interests in math, science, or technology could be attracted to K-12 teaching by h igher starting salaries, but that interest and ability factors limit the attractive effects of higher pay. The focus group results suggest that math, science, and technology majors see teach ing as a low paid field, that pay level was a significant factor making a career as a K-12 math or science teacher less attractive, and that many would consider it if it paid substant ially more than their current occupational choice. These results also show that students have a variety of reasons for not being attracted to teaching beyond low salaries. These in clude strong attachment to another career choice, doubts about their ability to be goo d teachers and discomfort with aspects of the job such as being responsible for others or standing in front of a class. For a significant minority of these students, even very l arge increases in entry pay are not likely to attract them to teaching. This was illustrated b y one focus group member who claimed she would not be interested in teaching even by a s alary 50% higher than that she

PAGE 16

16 of 25 expected from her relatively low paying field of as tronomy. The survey results suggest that the entry salaries for math and science teachers would not have to be raised to the same levels as engineering computer science, or the higher-paid health occupations to attract some of these student s. But the increases have to be greater than 5-10% to attract a substantial proportion of t hem. A beginning salary increase of about 25% would be needed to attract about 20% of t he respondents. The amount of increase does differ by student major, with higher increases needed to attract more engineering students than pure and applied science students. This is likely due to the higher salaries the former are expecting to receive in an engineering career. As expected, the entry salary level at which these students would consider a teaching career was related to the entry salary they thought they would receive in the non-teaching occupation they intended to enter. Evidence for the influence of personality traits and work values on the salary level at which these students would be willing to consider a teaching career was mixed. Focus group results suggested tha t personality and interest factors were important for some students in determining the attractiveness of K-12 teaching, and implied that a compensating differential (a salary level above what was expected for their current occupational choice) would be needed to att ract a substantial minority of these students. Survey results did not, however, find a c onsistent effect of personality factors and work values on the salary level at which respon dents were willing to consider a K-12 teaching career. Other factors that besides low pay did appear to re duce the attractiveness of a career in K-12 teaching. Focus group participants identified difficulties in dealing with children, taking work home, perceived intellectual monotony, and lack of up-to-date equipment in schools as unattractive aspects of teaching, but th ese were not as frequently expressed as concerns about lack of ability or interest in teach ing. Survey participants rated whether they would like or be good at teaching as more impo rtant considerations in deciding to pursue a K-12 teaching career were the pay equal to what they expected to receive in their chosen occupation than problems with student behavi or, lack of equipment, the extra work needed to get a license, and low status. While it m ay be that many of the study participants did not know enough about K-12 teaching to identify specific unattractive factors, these results suggest that interests and abilities might be more important than working conditions factors in influencing the choice of teaching as a career if salaries were increased. There are three implications of this study that tho se interested in attracting math, science, and technology majors to K-12 teaching by raising e ntry salaries may want to consider. First, significant increases in entry salary (e.g. 25%) would be needed to attract a substantial proportion of these students, though en try salaries would not have to be as high as in many alternative careers. Interests, per sonality, and values do not completely determine who will teach and do not completely rest rict the supply of labor. Second, it would make sense to target majors in fields that ar e not as highly paid, or where students are not expecting as high entry salaries. In this s ample, this would include pure science and heath-related majors. It is probably less effic ient to target engineering students, who in this sample on the average had higher pay expectati ons. Third, it is important to remember that some students are not going to be attracted to K-12 teaching by the higher salary levels that realistically could be implemented. Som e of these students would require a premium over what they expect to earn in their curr ent career choice to consider teaching, due to commitments to another career and concerns a bout their ability to teach. These are not likely to be worth trying to attract, both beca use of the potential need to pay much higher salaries and because these students may not make good teachers. While common sense still suggests that targeting th ose with personalities, interests, and

PAGE 17

17 of 25 values similar to those of students who have chosen teaching would be productive, this research did not find a consistent effect of these factors on attractive pay level. While the sophomore students in the survey sample with a stro nger value of concern for others indicated they would consider teaching at a lower e ntry salary, this relationship did not show up in the junior sample. The lack of a relatio nship between career risk aversion and attractive salary does not suggest that emphasizing the job security of K-12 teaching would be promising. LimitationsThe major limitation of this study is the uncertain generalizability of the results to all U.S. math, science, and technology majors due to the use of a sample from only one university and one point in time. The studies’ participants co nstitute only a small sample of those college students in the U.S. who have the knowledge base to teach science and math. Students at different universities in different par ts of the country may face different career options, different teacher beginning salaries, and have received different socialization. The research was also conducted before the dot com boom had bust, and in a period of economic expansion. Under different economic condit ions, some students would not be expecting such high beginning salaries and may see the potential job security of teaching as more attractive. The results are in need of repl ication before they can provide solid guidance to policy makers about pay levels needed t o attract students to math and science teaching.Another limitation is that the conclusions about th e exact size of the entry salary increase that would be needed to attract a substantial numbe r of students to K-12 teaching is influenced by the composition of the sample. The pr oportions of students in different majors influences the estimate of the proportion of students that would find teaching attractive at each salary level. Students with diff erent majors have different expected salary levels and in samples with a different mix o f majors this will result in different estimates of py levels needed to attract certain pe rcentages of these students. Though the slope of the ‘supply curve’ was fairly similar for engineering students, pure and applied science students, and the participant group as a wh ole, it may be better to try to estimate the pay-supply relationship separately for differen t majors, but in this study the sample size was not large enough to do so for any but the two m ost common majors. Ideally, one would also like to have much bigger sample sizes to ensur e a more trustworthy estimation of this relationship.There are also several limitations related to the m easures used. First, a more comprehensive measure of vocational interests might have shown a stronger relationship with attractive teaching salary than the dimensions from the Big 5 personality theory that were used. According to Dawis (1991), standard inte rest inventories are generally strong predictors of occupational choice, and might have h ad a stronger relation with attractive teaching pay. Second, the extroversion measure may not have been specific enough to represent the abilities required to be comfortably face the challenges of standing in front of a class or interacting with students that some of f ocus group participants identified as reasons for not teaching. It may be that more speci fic personality traits would have been more strongly related to the attractive teaching sa lary. Third, it is possible that, due to the length of the questionnaire, respondents became som ewhat careless in there responses to the work values and personality items. This migh t be the reason that these did not show a consistent relationship with attractive teaching salary. Lastly, the use of a limited range of categories of teaching beginning salaries restricte d the variance of the attractive salary variable. Had students been allowed to enter the do llar amount they would require to seriously consider K-12 teaching, the attractive sa lary would not have been censored from above and more information would have been availabl e for those in this study who chose

PAGE 18

18 of 25 the top category. Future ResearchTo provide more generalizable information about the beginning salary levels needed to attract students with knowledge and interests in ma th, science, and technology to K-12 teaching, this research would need to be repeated a t other universities. Improved and additional measures of personality and interests co uld be added. It might also be useful to explore more thoroughly the non-monetary reasons pe ople might not want to teach, and how they trade-off these factors with pay. Characte ristics like low status, lack of up-to-date technology, and an image of teaching as boring or a n intellectual dead end could be changed, and it might be interesting to see if doin g so would attract students more or less effectively as higher pay. One approach might be to present students with a series of scenarios which varied salary and other characteris tics in a classic ‘policy capturing’ design, in order to assess the relative importance of higher salaries and other conditions in attracting interest in teaching.AcknowledgmentThe assistance of Linda Smith Brothers of the Unive rsity of Wisconsin-Madison Graduate School of Business in conceptualization and data co llection for this project is gratefully acknowledged.Notes1. An earlier version of this paper was presented a t the American Education Finance Association annual conference held March 7-9, 2002 in Albuquerque, New Mexico. The research reported in this paper was partially suppo rted by a grant from the Carnegie Corporation of New York to the Consortium for Polic y Research in Education (CPRE) and the Wisconsin Center for Education Research, School of Education, University of Wisconsin-Madison (Grant No. B7136). The opinions e xpressed are those of the authors and do not necessarily reflect the view of the inst itutional partners of CPRE, the Carnegie Corporation of New York or the Wisconsin Center for Education Research. 2. Because the attractive salary response options e nded at $46,500, there is a ceiling effect on the relationship. The attractive salaries are censored from above. There is also a floor effect (or censoring from below) because the lowest attractive salary response option presented was $25,500. This is clear in the sophomo re group, but somewhat obscured in the junior group by two apparent outliers for which expected salaries are low, but attractive salaries are at the upper limit.ReferencesAmerican Federation of Teachers (1998). Survey and Analysis of Teacher Salary Trends 1998: AFT Teacher Shortage Survey Washington, DC: Author. Available at www.aft.org/research/survey/teacher.html.American Federation of Teachers (2001). Survey and Analysis of Teacher Salary Trends 2000 Washington, DC: Author. Available at www.aft.org/ research/. Ben-Shem, I., and Avi-Itzah, T.E. (1991). On work v alues and career choice in Freshman students: The case of the helping professions. Journal of Vocational Behavior, 39 369-379.

PAGE 19

19 of 25 Bradley, K. (1983). Recruitment to the teaching pro fession. Educational Research, 3 (2), 116-124.Cable, D.M., and Judge, T.A.(1994). Pay preferences and job search decisions: A person-organization fit perspective. Personnel Psychology, 47 317-348. Curan, B. Abrahams, C., and Manuel, J. (2000). Teacher Supply and Demand: Is There a Shortage ? Washington, DC: National Governors’ Association. Available at www.nga.org/Pubs/IssueBriefs/2000/Darling-Hammond, L., Berry, B.T., Hasselkorn, D., a nd Fideler, E. (1999). Teacher recruitment, selection, and induction: Policy Influ ences on the supply and quality of teachers. In L. Darling-Hammond and G. Sykes (eds.) Teaching as the Learning Profession : Handbook of Policy and Practices San Francisco: Jossey-Bass, 186-231. Dawis, R. V. (1991). Vocational Interests, values, and preferences. In M.D. Dunnette and L.M. Hough (eds.) Handbook of Industrial and Organizational Psycholog y 2nd Ed. Palo Alto, CA: Consulting Psychologist Press, 833-869.Eastin, D. (2000). California strives to end teache r flux. State Education Leader 18,2, 18-21. (Publication of the Education Commission of the States) Fetler, M (1997). Where have all the teachers gone? Education Policy Analysis Archives, 5 (2). Retrieved December 24, 2003 from http://epaa.a su/epaa/v5n2.html. Ferris, J. and Winkler, D. (1988). Teacher compensa tion and the supply of teachers. The Elementary School Journal, 86 (4), 389-403. Hare, D., Heap, J, and Raack, L. ( 2001, June). Tea cher recruitment and retention strategies in the Midwest: What are they and do the y work? NCREL Policy Issues, 8 1-6. Hogan, R., Hogan, J., and Roberts, B.W. (1996). Per sonality measurement and employment decisions: Questions and answers. American Psychologist, 51 (5), 469-477. Holland, J.L. (1985). Making Vocational Choices Englewood Cliffs, NJ: Prentice-Hall. Hirsh, E. (2001). Teacher Recruitment: Staffing Classrooms with Quali ty Teachers Denver, CO: State Higher Education Executive Office rs. Judge, T. A., Thoresen, C. J., Pucik, V., & Welbour ne, T. M. (1999). Managerial coping with organizational change: A dispositional perspec tive. Journal of Applied Psychology, 84 107-122.Meglino, B.M., and Ravlin, E.C. (1998). Individual values in organizations: Concepts, controversies, and research. Journal of Management, 24 (3), 351-389. Mount, M.K., and Barrick, M.R. (1995). The Big 5 pe rsonality dimensions: Implications for research and practice in human resource management. In Ferris, G.R, and Rowland, K.M. (Eds.) Research on Personnel and Human Resource Management 13, Greenwich, CT: JAI Press,153-200.Murnane, R.J. Singer, J.D., Willett, J.B., Kemple J.J., and Olsen, R.J. (1991). Who Will Teach? Policies that Matter Cambridge, MA: Harvard University Press. National Association of Colleges and Employers (200 1. Salary Survey: A Study of

PAGE 20

20 of 25 2001-2002 Beginning Offers 40, 2. Bethlehem, PA: Author. National Association of State Boards of Education ( 1998). The Numbers Game: Ensuring Quantity and Quality in the Teaching Workforce. The Reports of the NASBE Study Group on Teacher Development, Supply, and Demand. Alexand ria, VA: author. National Commission on Mathematics and Science Teac hing for the 21st Century (2000). Before It’s Too Late : A report from the National Commission on Mathematic s and Science Teaching for the 21st Century Washington, DC: U.S. Department of Education, Edu cation Publications Center.Pullen, S. (1999). Economic and policy determinants of science teacher supply. Journal of Research in Science Teaching, 35 (7), 745-755. Raaijmakers, Q. A. W. (1999). Effectiveness of diff erent missing data treatments in surveys with Likert-type data: Introducing the rela tive mean substitution approach. Educational and Psychological Measurement, 59 725-748. Recruiting New Teachers (2000). The Urban Teacher Challenge Belmont, MA: Author. Available at www.rnt.org.Ravlin, E.C., and Meglino, B. M. (1987). Effect of values on perception and decision-making: A study of alternative work values measures. Journal of Applied Psychology, 72 (4), 666-673. Saucier, G. (1994). Mini-markers: A brief version o f Goldberg’s unipolar Big 5 markers. Journal of Personality Assessment, 63 506-516. Tokar, D.M., and Swanson, J.L. (1995). Evaluation o f the correspondence between Holland’s vocational personality typology and the f ive-factor model of personality. Journal of Vocational Behavior, 46 89-108. Tusin, L.F., (1999). Deciding to teach. In R.P. Lip ka and T.M. Brinthaupt, (Eds.) The Role of Self in Teacher Development. Albany, NY: SUNY Pr ess, 11-35.About the AuthorAnthony MilanowskiConsortium for Policy Research In EducationWisconsin Center for Education ResearchIndustrial Relations Research InstituteUniversity of Wisconsin-MadisonMadison, WI 53706(608) 263-4260Email: amilanow@facstaff.wisc.eduAnthony Milanowski is an Associate Researcher with the Consortium for Policy Research in Education (CPRE) and a participating faculty mem ber of the Industrial Relations Research Institute at the University of Wisconsin-M adison. He received his Ph.D. in Industrial Relations from the University of Wiscons in-Madison in 1997. Prior to joining CPRE, he worked in human resource management for 16 years, primarily with the Wisconsin Department of Employment Relations. His c urrent research interests include teacher performance evaluation, pay system innovati ons, and the teacher labor market.

PAGE 21

21 of 25 Appendix Table 1 Beginning Salary Offers for Selected Occupations fr om Spring 2001 National Association of Colleges and Employers Sala ry SurveyMajorMedian Offers,Number of OffersMathematics 46,000197 Chemistry40,00091Biological & Life Sciences27,600320Physics52,00049Engineering – Civil40,3001184Engineering – Chemical52,0001305EngineeringElectrical/Electronic 52,0001883EngineeringMechanical49,0002225Computer Science52,1001428Accounting40,0001807Elementary Education30,000480Secondary Education28,700204Appendix Table 2 Median Expected Annual Beginning Salaries and Attra ctive Teaching Salaries, Students Interested in Math and Science, and Technology MajorsSample and MajorNumberExpected Salary Attractive Salary Attractive 12 Month = SophomoresBusiness (e.g. Accounting)637,50036,50043,800Engineering1842,50040,50048,600Health Related2038,75038,50046,200Math & Computer652,50045,50054,600Science5240,00039,50047,400All10240,00040,50048,600JuniorsBusiness (e.g. Accounting)150,00046,500+55,800Engineering4245,00041,50049,800Health Related1335,00040,50048,600Math & Computer1450,00040,50048,600

PAGE 22

22 of 25 Science3640,00040,50048,600All10645,00040,50048,600Appendix Table 3 Means, Standard Deviations, and Correlations of Var iables Used in Tobit Analyses, Sophomore SampleN=102 Variable MSD 123456789 1. Natural Log, Attractive Salary 10.54630.1821 2. Natural Log, Expected Salary 10.66850.4674.414 3. Female0.5980.493-.175-.1504. Intend to Attend Grad School 0.6470.480.126.172.022 5. Grade Point Averagea5.201.851.092.091-.010.268 6. Career Risk Aversion 2.510.499-.047.236.010-.019.048 7. Extroversion10.40611.238-.022-.048.216-.054.095.199 8. Openness19.5367.799.059-.018-.071.103-.005-.218. 045 9. Agreeableness 21.0497.881-.227-.153.344-.074.137-.152.433.146 10. Concern for Others 6.4612.228-.278-.078.017-.078.170.196-.049-.058.290 (a) This is the average of the ordered categories c hosen. It corresponded to the GPA category of 3.0-3.2.Appendix Table 4 Means, Standard Deviations, and Correlations of Var iables Used in Tobit Analyses, Junior SampleN=105 Variable MSD 123456789 1. Natural Log, Attractive Salary 10.58770.1584

PAGE 23

23 of 25 2. Natural Log, Expected Salary 10.71030.3435.276 3. Female0.371 0.486-.008-.3254. Intend to Attend Grad School 0.4290.497.174.204.171 5. Grade Point Averagea5.2901.763.134.151.077.167 6. Career Risk Aversion 2.4450.536-.075-.034.045-.157.070 7. Extroversion5.87611.655.022.068.124.134-.054-.15 2 8. Openness19.3338.277-.065.074-.024.044-.026-.282. 347 9. Agreeableness 20.0198.296-.106-.047.225.014.151.128.306.222 10. Concern for Others 5.1902.232-.072-.120-.013-.152.008.350-.036-.123.17 2 (a) This is the average of the ordered categories c hosen. It corresponded to the GPA category of 3.0-3.2. The World Wide Web address for the Education Policy Analysis Archives is epaa.asu.edu Editor: Gene V Glass, Arizona State UniversityProduction Assistant: Chris Murrell, Arizona State University General questions about appropriateness of topics o r particular articles may be addressed to the Editor, Gene V Glass, glass@asu.edu or reach him at College of Education, Arizona State Un iversity, Tempe, AZ 85287-2411. The Commentary Editor is Casey D. Cobb: casey.cobb@unh.edu .EPAA Editorial Board Michael W. Apple University of Wisconsin David C. Berliner Arizona State University Greg Camilli Rutgers University Linda Darling-Hammond Stanford University Sherman Dorn University of South Florida Mark E. Fetler California Commission on TeacherCredentialing Gustavo E. Fischman Arizona State Univeristy Richard Garlikov Birmingham, Alabama Thomas F. Green Syracuse University Aimee Howley Ohio University

PAGE 24

24 of 25 Craig B. Howley Appalachia Educational Laboratory William Hunter University of Ontario Institute ofTechnology Patricia Fey Jarvis Seattle, Washington Daniel Kalls Ume University Benjamin Levin University of Manitoba Thomas Mauhs-Pugh Green Mountain College Les McLean University of Toronto Heinrich Mintrop University of California, Los Angeles Michele Moses Arizona State University Gary Orfield Harvard University Anthony G. Rud Jr. Purdue University Jay Paredes Scribner University of Missouri Michael Scriven University of Auckland Lorrie A. Shepard University of Colorado, Boulder Robert E. Stake University of Illinois—UC Kevin Welner University of Colorado, Boulder Terrence G. Wiley Arizona State University John Willinsky University of British ColumbiaEPAA Spanish and Portuguese Language Editorial BoardAssociate Editors for Spanish & Portuguese Gustavo E. Fischman Arizona State Universityfischman@asu.eduPablo Gentili Laboratrio de Polticas Pblicas Universidade do Estado do Rio de Janeiro pablo@lpp-uerj.netFounding Associate Editor for Spanish Language (199 8-2003) Roberto Rodrguez Gmez Universidad Nacional Autnoma de Mxico Adrin Acosta (Mxico) Universidad de Guadalajaraadrianacosta@compuserve.com J. Flix Angulo Rasco (Spain) Universidad de Cdizfelix.angulo@uca.es Teresa Bracho (Mxico) Centro de Investigacin y DocenciaEconmica-CIDEbracho dis1.cide.mx Alejandro Canales (Mxico) Universidad Nacional Autnoma deMxicocanalesa@servidor.unam.mx Ursula Casanova (U.S.A.) Arizona State Universitycasanova@asu.edu Jos Contreras Domingo Universitat de Barcelona Jose.Contreras@doe.d5.ub.es

PAGE 25

25 of 25 Erwin Epstein (U.S.A.) Loyola University of ChicagoEepstein@luc.edu Josu Gonzlez (U.S.A.) Arizona State Universityjosue@asu.edu Rollin Kent (Mxico) Universidad Autnoma de Puebla rkent@puebla.megared.net.mx Mara Beatriz Luce (Brazil) Universidad Federal de Rio Grande do Sul-UFRGSlucemb@orion.ufrgs.br Javier Mendoza Rojas (Mxico)Universidad Nacional Autnoma deMxicojaviermr@servidor.unam.mx Marcela Mollis (Argentina)Universidad de Buenos Airesmmollis@filo.uba.ar Humberto Muoz Garca (Mxico) Universidad Nacional Autnoma deMxicohumberto@servidor.unam.mx Angel Ignacio Prez Gmez (Spain)Universidad de Mlagaaiperez@uma.es DanielSchugurensky (Argentina-Canad) OISE/UT, Canadadschugurensky@oise.utoronto.ca Simon Schwartzman (Brazil) American Institutes forResesarch–Brazil (AIRBrasil) simon@sman.com.br Jurjo Torres Santom (Spain) Universidad de A Coruajurjo@udc.es Carlos Alberto Torres (U.S.A.) University of California, Los Angelestorres@gseisucla.edu EPAA is published by the Education Policy Studies Laboratory, Arizona State University