Issue No. 9
Fall 2004

Briefs & Abstracts
A Review of Published Research & News

Willard Hom

Professional workers in the community colleges bear a weighty workload these days. Quite often their bustling routines force them to focus on specific topics of immediate concern, and this regimen leaves them little time to scan the vast universe of research and news for tidbits of information that could broaden their knowledge base for future work or offer them new insights for current work. This section of the journal is our attempt to simulate a visit to the college library where an inquiring mind, given some time, might browse for topics that pique one’s professional interest. And to acknowledge modern times, we even include browsings from the virtual library of the Internet.

We use abstracts of published research to provide readers with chunks of information. These chunks often include quotations from the original articles so that readers can directly employ them for their own purposes (presumably noble ones). To help readers evaluate the applicability of the abstracted work, we include some notes about each study’s staff, data, and methods, as well as a citation or further reading. These abstracts usually occupy only one page so that readers can capture the essential points quickly. The brevity also promotes the sharing of this knowledge with others; we encourage readers to disseminate abstracts of their choice. Although the entire set of abstracts in this issue came from the pen of one person, we hope that others will soon contribute abstracts to this section. Just submit a Word document to me at whom@cccco.edu

The section ends with a select listing of web pages that readers may want to explore on their own. Again, we appreciate the referral of new and interesting web pages that readers may encounter. Just e-mail me at the above address.

Table of Contents

Click on the article you wish to examine or scroll down to see them all.

Student Learning/Participation

1. A Qualitative Study Exploring Black Student Success and Family Ties
   (How does family relate to the academic achievement of black college students?)
2. Course-taking Patterns Among Community College Students
   (Are there meaningful patterns in the course-taking behavior of entering community college students?)
3. The Link Between Attitude Toward Math and Math Achievement
   (Does attitude of K-12 students toward math determine their success in math, or vice-versa?)
4. Community Colleges as a Bridge to Science & Engineering Degrees
   (What role do community colleges play in the education of the nation’s scientists and engineers?)
5. A Meta-Analysis on Self-Beliefs and Academic Achievement
   (For K-12 students, how much, if any, linkage is there between self-beliefs and later academic achievement?)
6. Meeting Community College Student Needs
   (What kinds of administrative changes might community colleges make to improve the success of at-risk students?)

Teaching & Faculty

1. A Downward Trend in Teacher Quality?
   (In K-12, is there a decline in the quality of teachers?)

Workforce Development

1. International Context of Nursing Supply and Demand
   (What situations in other countries affect the need for domestic production of a pool of nurses for the United States?)
2. Wages: An Incomplete Measure of Labor Outcomes
   (Is the wage level for an individual the one and only yardstick for the quality of labor force outcomes?)

Strategic Planning

1. About Types of Rapid County Population Growth
   (Does California have places with unique rates of population growth?)

Research Methods

1. A Different Approach for Analyzing Transfers from 2-Year Colleges
   (How can transfer research progress without jeopardizing community colleges?)
2. On Reporting the Effect Size of a Study
   (Does it matter how a research report states a statistical finding?)
3. New Evidence on the “No Opinion” Response Option
   (In survey question design, does it help to include the option for “no opinion?”)

Links to Other Publications of Interest

A Qualitative Study Exploring Black Student Success and Family Ties

Two researchers recently published a qualitative study on the influence of family ties to the academic success of black college students. Among their findings are the following:

1. “...families lay the groundwork for success long before Black students get to college. In this study, the term family included extended family like grandparents, aunts, and uncles, and fictive kin (neighbors, church members, friends). Black families that aspire to send their children to college, then, need to establish expectations with respect to education early in their children’s lives...fictive kin can play a critical role...” [p.505]
   
   
2. “Family and fictive kin can also be aware of the influence they have on their children in terms of perspectives on race. The way they talk with their children about race and the experiences they and their offspring have with race all influence children when it comes to college. Whatever steps parents and other relatives [and fictive kin] can take to help children understand how to succeed as a minority in a majority culture will help those children as they persist in education...” [pp. 505-6]
   
   
3. “What motivates precollege students is having a role model or serving as a role model. Family and fictive kin can help children by holding them up as models to other children and introducing them to others who have succeeded in education. It is also important to note the lack of male role models for young African Americans. All but one of the family members in this study were female. Moreover, student participants routinely reported the influence of females in their lives but did not report the same for males...” [p.506]
   
   
4. “...current and African American students and Black alumni...understand the rigors of postsecondary education—both the personal and academic challenges and rewards..As the evidence in the study suggests, many students are already serving as mentors for siblings and cousins. They might expand their role, however, and develop relationships with other younger children who might not have a role model with respect to higher education...” [p.506]
   
   
5. “...those who work to recruit Blacks to college...need to work with whole families to promote college attendance. This means introducing programs for families of elementary and secondary school children so that precollege events and experiences are purposefully introduced. Such programs would assist families in instilling educational values and aspirations in their children. Families might also benefit from discussions about race and how to deal with discrimination. These programs might also offer advice on how to motivate young Blacks to aspire to higher education and teach children about the doors that a college degree can open...” [pp. 506-7]
   
   
6. “With respect to the second stage in the model [that the researchers derived from the data], early college...families and fictive kin...need to make sure their students understand how minority students in a predominately White university can survive in both the academic and social realms...family members need to familiarize themselves with the resources available to their students so that they can encourage their students to use those resources should difficulties arise...The data suggest that having support networks on campus creates a sense of community for Black students...” [p.507]
   
   
7. “...spiritual support is important to student success. Families can make sure their students have a religious organization with which to affiliate while at school...” [p.507]
   
   
8. “Current Black students and alumni can also take action to assist others in the early college stage...adopting a little brother or sister for whom they can serve as a resource about navigating the academic and social milieus on campus...” [p. 507]
   
   
9. [For campus administrators] “...certain additional measures might assist African American students in succeeding once they matriculate. Actively helping students negotiate the new environment, for example, might involve offering seminars for small groups of Black students in which the problems they encounter are discussed and the successes they achieved are acknowledged. Such programs might also promote social networks among students and between students and staff...administrators might offer more assistance in the area of spiritual support...” [p.508]
   
   
10. [In the late college stage of black college student life], “...successful students feel an obligation to their families and believe their personal triumphs are triumphs for their families, as well. They believe they need to repay their families for the support they have received. For some, this debt may be repaid by serving as a role model for younger members of the family and mentoring others through the educational process...Professionals [campus staff] might also establish programs that connect recent alumni to upper division students. Doing so would provide a mechanism for alumni to serve as role models and meet their own expectations of repayment. Such programs would also be self-promoting. Alumni would work with upper division students who would soon become alumni and might, therefore, be more likely to mentor upcoming upper division students...” [p.509]

These findings, if valid in other states, have much relevance for four-year institutions that strive to improve the success of their Black students. However, two-year institutions can benefit by recognizing these aspects within the “bridging” role that it can perform for Black students who transfer from their campuses to four-year campuses. In addition, the proximity of two-year campuses to the communities of Black secondary students gives them an immediate advantage in helping in the pre-college stage (which would presumably include extensive outreach efforts in conjunction with four-year institutions). Finally, the study demonstrates the kind of information that a qualitative study can generate for administrators/staff involved with student services. (Granted, many student services staff have already tacitly understood these issues, but the study may help to corroborate their intuitive knowledge based on field work.)

Michael K. Herndon (director of the Interdisciplinary Studies undergraduate degree program at Virginia Tech) and Joan B. Hirt (associate professor of higher education and student affairs at Virginia Tech) document their study in an article (“Black Students and Their Families: What Leads to Success in College) published by the Journal of Black Studies (March 2004, Volume 34, No.4, pp. 489-513). Along with 33 references and a discussion of study limitations, they detail their qualitative methodology. The study used in-depth interviews of 20 Black students (11 females and 9 males) and their families. The students were seniors at two predominately White universities in a mid-Atlantic state.

[Abstract done by Willard Hom, Director, Research & Planning Unit, System Office, California Community Colleges, 8/24/04]

Course-Taking Patterns Among Community College Students

A recently published article describes how a team of researchers explored the oft-ignored aspect of course-taking patterns among entering (or “first-time”) community college students. The team used data on 13,108 first-time students who had registered for courses in the Los Angeles Community College District (LACCD) for the spring academic term. In the team’s words, “our goal is to identify whether or not student patterns emerge when using transcripts and courses as a unit of analysis.” [p.22] Among the team’s findings are the following:

1. “The dispersal of the entering students is a dramatic feature of the colleges. First-time students are not concentrated in a few main introductory courses...[This]...dispersal of first-time students is one of the distinguishing, but often unnoted, characteristics of the community college and possibly one of several reasons why there is so little community among the students. There is no broad common first year, or even first semester, experience for the new students...” [p.39]
   
   
2. “The classification of courses as academic, occupational, or ESL, and as remedial, A.A. applicable/not transferable, or A.A. applicable/transferable levels, was found to be useful for identifying relationships with the student characteristics of gender, ethnicity, age, and full-time or part-time enrollment...” [p.40]
   
   
3. “There was not a clear pattern of tracking and exclusion in vocational and academic courses along a white/non-white divide. White students were interested in most of the vocational courses examined here. East Asians [ ] were the most likely of all ethnic groups to enroll in a vocational area such as computer science...” [p.40]
   
   
4. “In contrast to the younger students, the students over the age of 24 were more likely to enroll in occupational courses...” [p.41]
   
   
5. “Counselors and curriculum administrators may be concerned that large proportions of the students in the remedial English or mathematics courses are taking four or more courses, and that well over one-half are enrolled in three or more courses...” [p.41]
   
   
6. “Lack of language skills creates one of the great divides in the community college. Thus in the beginning ESL courses that were examined, the students displayed very different features from the other first-time students. It was only in these ESL courses that first-timers were in the majority. The majority of the students were white despite their being a minority in the general student population...” [pp.41-42]

This derivative of the TRUCCS (Transfer and Retention of Urban Community College) research project demonstrates some of the potential for analyzing transcript data of community college students along with course-level enrollment composition. The uniqueness of the LACCD student sample may limit the generalizability of the explicit patterns discovered but the model of data and analysis should still have transferability to many other environments (including colleges outside of California).

The team of researchers behind this article were William Maxwell; Linda Serra Hagedorn; Scott Cypers; Hye S. Moon; Phillip Brocato; Kelly Wahl; and George Prather. Maxwell and Serra Hagedorn are faculty in the School of Education at the University of Southern California and leaders in the TRUCCS project. Cypers, Moon, and Brocato are staff in TRUCCS. Wahl is director of assessment and data analysis at Loyola Marymount University while Prather is chief of the office of institutional research for the LACCD. Interested readers may find their article (“Community and Diversity in Urban Community Colleges: Coursetaking Among Entering Students”) in the refereed journal, Community College Review (vol. 30, n.4, pp. 21-45). The article includes 31 references, some literature review, study limitations, and future needs.

[Abstract by Willard Hom, Director Research & Planning, System Office California Community Colleges, completed 8/17/04]

The Link Between Attitude Toward Math and Math Achievement

Two researchers, Ma & Xu, recently explored the common belief that secondary-level students must have a positive attitude towards math in order to succeed in a math course. Applying a statistical method known as structural equation modeling, they concluded the following:

1. “...prior achievement significantly predicted later attitude across grades 7-12. Prior attitude, by contrast, did not meaningfully predict later achievement. We conclude that achievement demonstrated causal predominance (priority) over attitude in the entire secondary school...” [p. 273]
   
   
2. “We found no evidence of gender differences in the causal relationship between attitude toward mathematics and achievement in mathematics...” [p. 273]
   
   
3. “Elite status in mathematics (taking calculus, the most advanced course in mathematics) did moderate the causal relationship between attitude toward mathematics and achievement in mathematics...We found an imbalanced reciprocal relationship between attitude and achievement across almost secondary school [sic] for nonelite students, with achievement showing causal predominance over attitude. Such a reciprocal relationship was not found among elite students. For these students, when there was a causal relationship between attitude and achievement, achievement always claimed (unidirectional) causal predominance over attitude...” [p. 274]
   
   
4. “...late junior and early senior high school grades are the most effective periods to use achievement to promote attitude (coinciding with the crucial period for students’ attitudinal formation). Interventions that are implemented either sooner or later may not adequately achieve their desired purposes...We suggest that nonelite students in mathematics can be especially responsive to this strategy...an initial effort to improve attitude (in late junior or early senior high school grades) can have a far-reaching impact into the circle of attitude and achievement...” [p. 277]

In terms of data, Ma & Xu used six waves of data (grades 7-12) from the Longitudinal Study of American Youth (LSAY). This panel survey tracks secondary students from a nationally representative sample, giving the study 3,116 students for the analysis (1,626 males and 1,490 females). The data analysis used three survey indicators for attitudes toward math (capturing the construct of utility of mathematics) and four indicators for math achievement (basic skills, algebra, geometry, and quantitative literacy). They also noted the limitations in their study. The measurement of attitude toward math omits the frequently considered attitudinal components of confidence and enjoyment in math. The quasi-experimental design of the study cannot confirm a causal linkage, and more research is needed to achieve confirmation.

This study has relevance for researchers in higher education. If this study is valid for postsecondary students, the institution’s ability to affect their level of achievement in math via attitudinal change is perhaps minimal. The lack of a gender effect and the discovery of an “elite” effect should also be of interest among higher education researchers.

Xin Ma (associate professor of mathematics education, Dept. of Curriculum and Instruction, University of Kentucky) and Jiangmin Xu (Ph.D. Sociology, Brigham Young University) document their data, methods, limitations, and literature review in their article (“Determining the Causal Ordering between Attitude Toward Mathematics and Achievement in Mathematics”) in American Journal of Education (May 2004 issue, volume 110, pp. 256-280). The article includes four tables and 31 references.

[Abstract done by Willard Hom, Director, Research & Planning Unit, System Office, California Community Colleges, 8/20/04]

Community Colleges as a Bridge to Science & Engineering Degrees

An April 2004 report from the National Science Foundation (NSF) provides some strong evidence about the contribution that the nation’s community colleges have made to the production of postsecondary degrees in the sciences. Using data from the NSF’s 2001 National Survey of Recent College Graduates, the report presented the following findings, among others:

1. “On average, 44 percent of S&E [science & engineering] graduates attended community colleges...” [p.1]
   
   
2. “[From responses to the 1999 survey,] the two most important reasons for attending a community college were to complete credits toward a bachelor’s degree (74 percent) and to gain further skills and knowledge in an academic or occupational field (50 percent). Earning an associate’s degree ranked sixth out of the nine reasons ranked...Less than 30 percent of S&E graduates who have attended community college earn an associate’s degree...” [p.2]
   
   
3. “Hispanic S&E graduates are more likely to have attended a community college than any other racial/ethnic group. About 51 percent of Hispanic S&E graduates reported attending community college before receiving their bachelor’s or master’s degree, compared with 45 percent of American Indians/Alaskan Natives, 44 percent of blacks, 43 percent of whites, and 40 percent of Asians/Pacific Islanders...” [p.2]
   
   
4. “American Indian/Alaskan Native and Hispanic doctorate holders are more likely to have attended community college than Asian/Pacific Islander, black, or white doctorate holders...Among Hispanic doctorate holders, Mexican Americans are the most likely to have attended community college, 18 percent of whom had attended community college before receiving their doctoral degrees...” [pp.2-3]
   
   
5. “...72 percent of 1999 and 2000 S&E graduates 50 years of age and older had attended community college, compared with 32 percent of S&E graduates 24 years of age and younger...” [p.3]
   
   
6. “S&E graduates who have been divorced, widowed, or separated are more likely to have attended community college than S&E graduates who either are married or have never been married. Only 39 percent of graduates who have never been married attended a community college, compared with 76 percent of divorced graduates...Among female [S&E] graduates, 62 percent with children in the household had attended a community college, compared with 44 percent of those without children in the household...” [p.3]
   
   
7. “About 57 percent of the S&E graduates who reported that their fathers had less than a high school diploma had attended a community college, whereas only 36 percent of those reporting that their fathers had some graduate or professional school had attended community college...” [p.4]
   
   
8. “S&E graduates who attended high school in...Alaska, California, Hawaii, Oregon, Washington, and outlying areas [of the Pacific region], are more likely to have attended a community college than are graduates in other regions of the United States. This may be a reflection of California’s extensive community college system...[p.4]

These findings buttress the position that community colleges act as an important bridge that diverse populations use to achieve postsecondary degrees in science and engineering. We also find evidence that many students forego the associate’s degree in their path to a four-year degree (making it misleading to use the volume of associate’s degrees as a stand-alone measure of community college performance). John Tsapogas of the NSF authored the report ("The Role of Community Colleges in the Education of Recent Science and Engineering Graduates," Report NSF 04-315), available at:
http://www.nsf.gov/sbe/srs/infbrief/nsf04315/start.htm

[Abstract by Willard Hom, Director, Research & Planning Unit, System Office, California Community Colleges, May 15, 2004]

A Meta-Analysis on Self-Beliefs and Academic Achievement

Do students’ self-beliefs affect their academic achievements? This important question in education received a critical test from a trio of researchers (Valentine, DuBois, and Cooper—or VBC) who summarized the relevant research on this topic (55 studies that appeared from 1978-2001) through the modern statistical approach known as meta-analysis. Using statistics from 60 independent samples in these studies (covering more than 50,000 K-12 students), VBC made the following conclusions, among others:

1. “Overall, available findings are consistent with the view that self-beliefs can influence academic achievement. The magnitude of the overall estimated relation between self-beliefs and later achievement, controlling for initial achievement, is not large (beta = .08), but does approach Cohen’s (1988) threshold of r = .10 for a small effect size in the social sciences...” [p.127]
   
   
2. “...among equally achieving students, having positive self-beliefs confers a small but noteworthy advantage on subsequent achievement measures relative to students who exhibit less favorable self-beliefs...The claim that self-beliefs are either entirely irrelevant to student achievement or likely to be detrimental in their effects...are not consistent with the cumulative evidence...” [p.127]
   
   
3. “...our results should not be taken as proof that self-beliefs have a causal relation to later achievement...” [p.128] [Comment: This is based on the quasi-experimental design used here.]
   
   
4. “...the findings of the meta-analysis are equally clear in suggesting that the effects of student beliefs on achievement typically are small in magnitude. Even allowing for methodological limitations, evidence is lacking to support theoretical or applied perspectives on which self-beliefs are characterized as strong and pervasive influence on student achievement...” [p.127]
   
   
5. “...The importance of specificity of measurement is further suggested by the evidence of stronger effects when assessments focus on self-beliefs and achievement for the same domain, such as a particular subject area in school...” [p.128] [Comment: an example of this would be self-beliefs about math ability and math achievement.]
   
   
6. “Issues of concern with respect to sampling characteristics include a lack of non-Western countries, those from varying ethnic and racial backgrounds, and those in the earliest stages of schooling...” [p.129] [Comment: This is one of the study’s limitations.]

This study has relevance to educational researchers because it clarifies the positive, but modest, link between student self-belief and academic achievement. This has implications for potential interventions to improve student achievement. In addition, the detailed coverage of the meta-analysis technique applied here may help other researchers who use these results and/or who may undertake their own meta-analyses. Jeffrey C. Valentine (Duke University), David L. DuBois (University of Illinois-Chicago), and Harris Cooper (Duke University) document their study in an article (“The Relation Between Self-Beliefs and Academic Achievement: A Meta-Analytic Review”) in the refereed journal, Educational Psychologist (Spring 2004, Vol.39, No.2, pp. 111-133). The article includes about 3 ½ pages of references. The journal editors note that requests for reprints should go to Professor Valentine at:

Jeffrey C. Valentine, Department of Psychology, Duke University, Box 90739, Durham, NC 65211. The e-mail address for requests is as follows: jeff.valentine@duke.edu

[Abstract done by Willard Hom, Director, Research & Planning Unit, System Office, California Community Colleges, 8/29/04]

Meeting Community College Student Needs

David Glenn reports in The Chronicle of Higher Education (Tuesday, August 17, 2004) on an interesting story about how public community colleges (CCs) could help financially needy, first-generation students. A trio of researchers from Northwestern University presented the results of a study of 14 two-year colleges (7 public and 7 private). Based primarily on interviews of 86 students and 96 administrators/faculty, they concluded the following:

“...several organizational structures that seemed to raise graduation rates...[are] one-stop shopping...predictable and streamlined curricula...and low counselor-student ratios...” [bold face added for emphasis]

The concept of one-stop shopping would address the concern students termed as “frustrating and confusing trips to several different offices.” The concept of predictable and streamlined curricula would address the desire for “courses offered in a regular sequence at convenient times of the day.” The concept of low counselor-student ratios would address the concern about registering for the wrong classes and missing opportunities.

The study noted that private two-year colleges seemed to address these three concepts more successfully than public two-year colleges did. Using a survey of 4,400 students from the 14 colleges, they found that 43 percent of the sample from the public sector felt that it would take a year longer than they expected to get the associate degree. In comparison, only 23 percent of the private sector sample gave that response. Of course, the researchers noted that public CCs have a wider set of missions than private two-year colleges have, and this situation—a mitigating circumstance for the observed differences between the two sectors.

Further details are in the original news story (“For Needy Students, College Success Depends no More Than Access, Study Finds”). The trio of researchers from Northwestern University are Ann E. Person (graduate research assistant), James E. Rosenbaum (professor of sociology and education), and Britt Gordon-McKeon (undergraduate research assistant). All three work in Northwestern’s Institute for Policy Research. [Note: This abstract relies upon the trio’s presentation at the recent annual meeting of the American Sociological Association, as reported by David Glenn, so I have no formal paper to which I can refer readers for further information.]

[Abstract done by Willard Hom, Director, Research & Planning Unit, System Office, California Community Colleges, 8/24/04]

A Downward Trend in Teacher Quality?

The efforts to improve the performance of our K-12 system has put some focus on the quality of the teachers in that system. A trio of researchers recently published a study that explores one dimension of teacher quality that has broad implications. Corcoran, Evans, & Schwab analyzed the change in the population who become teachers as a partial test for the hypothesis that the broadening of job opportunities for women has affected the pool of candidates who have recently begun careers in teaching. Among their salient points are the following:

1. “... a growing body of evidence...shows that certain measures of teacher quality—in particular, their verbal and mathematical skills—are strongly related to student outcomes. The persistence of this finding stands in sharp contrast to the continuing debate over the importance of other measured inputs into the production of education (notably, per-pupil expenditures and class size)...” [p.450]
   
   
2. “Over the 1964-2000 period, women near the top of the test score distribution became much less likely to enter the teaching profession than their peers near the middle of the distribution. The apparent consequence has been a much lower representation of women of very high academic ability in the pool of elementary and secondary teachers. While the sample sizes of male teachers are much smaller, we detect the opposite trend among men...” [p467]
   
   
3. “If our results can be applied to the wider population of young teachers in the United States, a given student in 2000 (conditional on having a female teacher) could expect to find a teacher who is on average of only slightly lower academic ability than a given student in 1964. However, that student is much less likely to find a teacher of the highest academic ability than a student in 1964...the likelihood that a student in a low income or predominately black school encounters a teacher of the highest academic ability is likely to be even lower...if the significant loss of women in the top decile—those who likely stood to benefit most from occupational desegregation—is indicative of a wider trend, then these findings should be of interest to parents, researchers, and policymakers alike...” [p.467]

The study has relevance to policymakers in that a critical input to the education process seems to demonstrate a noteworthy trend, the gradual loss of high-ability women into the teaching profession (and a possible increase in high-ability males into the profession does not seem sufficient to offset this loss). From a methodological viewpoint, this study also demonstrates the importance of analyzing the variation across the entire distribution of a variable (in this case the distribution of teacher quality) rather than analyzing highly aggregated data or the central portion of the distribution (and assuming a linear relationship).

The trio of Sean P. Corcoran (Dept. of Economics, California State University, Sacramento), William N. Evans (Dept. of Economics, University of Maryland, College Park), and Robert M. Schwab (Dept. of Economics, U.Maryland, College Park) describe their data, analytical methods, and the related literature in their article (“Women, the Labor Market, and the Declining Relative Quality of Teachers”) in the Journal of Policy Analysis and Management (Summer 2004, Vol.23, No.3, pp. 449-470). An appendix lists the seven data sources they used, and they include nearly three pages of references (potentially helpful for other researchers).

[Abstract by Willard Hom, Director Research & Planning, System Office California Community Colleges, completed 7/27/04]

International Context of Nursing Supply and Demand

The need for California, and the nation as a whole, to plan for increasing the output of nurses by its postsecondary institutions reflects international trends as well as domestic factors in supply and demand. In a pair of recent articles published in Health Affairs (a peer-reviewed journal) this international dimension becomes clearer.

The articles make the following points:

1. Although the demand for foreign nurses to fill jobs in the United States fluctuated over the past two decades, a noteworthy trend has begun. “Although the proportion of foreign nurses has never exceeded 5 percent of the U.S. nurse workforce, that figure is now slowly rising...In 1995 nearly 10,000 foreign nurses received their U.S. RN licenses, representing almost 10 percent of all newly licensed RNs in that year. By 1998 that proportion fell by nearly half, as the number of new foreign nurses entering the U.S. workforce fell more steeply than the number of new U.S. RNs...After 1998 the foreign nurse proportion steadily grew, topping 14 percent in 2003...”[p.79, A]
   
   
2. The reliance upon foreign sources for domestic nursing needs involves some uncertainties that may affect the continued use/expansion of foreign sources. “No studies to date have determined whether foreign nurses’ cultural orientations and technical competence produce differences in patient outcomes when compared with their domestic counterparts...” That is, should we continue to rely on foreign sources if research were to find that foreign nurses, on the average, produce less positive patient outcomes than do domestic nurses?
   
   
3. The continuity of foreign sources may change. For example, Ireland recently experienced an economic upturn. This resulted in that nation becoming an importer of nurses instead of an exporter; it now competes with the United States to recruit foreign nurses. The Philippines has provided the largest proportion of foreign nurses to the U.S. labor market. However, “it is estimated that there are more than 30,000 unfilled nursing positions in the Philippines.” [p.75, B] At some point, the Philippine government may begin to restrict the outflow of its nurses (although their work abroad has resulted in an influx of money to the nation—because of the tradition of Filipino nurses sending funds to support their families in the homeland). Recently, countries in Africa have exported nurses to the developed countries, somewhat increasing the supply of nurses for more developed countries. But these source countries suffer a toll. “The accelerated recruitment of experienced African nurses is straining an already fragile health care infrastructure in many African countries, which have been battered by AIDS and deprived resources because of economic and political upheaval...Receiving countries obtain the financial benefit of the migrant’s professional education and training, while sending countries bear the costs...” [pp.81-82, A]
   
   
4. A policy bottom-line here is that the United States, including California, must plan to make a substantial expansion in its capacity to produce new nurses. As a current major source of new nurses, the community college system could obviously play a role in such a plan if there were resources to expand its capacity. Potential nursing students are not lacking. “In 2003 more than 11,000 qualified students were turned away from U.S. nursing education schools because of capacity limitations. Unlike in most other countries, nursing education in the United States is largely financed by students and their families, a financial barrier for many applicants...” [p.76, B] As these articles note, “increasing demand for foreign nurses in the face of greater domestic production is a signal that domestic efforts are insufficient to keep up with demand..” [p.85, A]

For additional details, you may refer to the complete articles themselves:

A. “Imported Care: Recruiting Foreign Nurses to U.S. Health Care Facilities” in Health Affairs, May/June 2004, V.23, N.3. 78-87. The authors are Barbara L. Brush (associate professor in the William F. Connell School of Nursing at Boston College, in Chestnut Hill, Mass.), Julie Sochalski (associate professor in the School of Nursing at the University of Pennsylvania in Philadelphia), and Anne M. Berger (doctoral candidate in the Connell School of Nursing).

B. “Trends in International Nurse Migration” in Health Affairs, May/June 2004, V.23, N.3. 69-77. The authors are Linda H. Aiken (professor of sociology and director of the Center for Health Outcomes and Policy Research, School of Nursing, University of Pennsylvania in Philadelphia), James Buchan (professor, Queen Mary University College in Edinburgh, Scotland), Julie Sochalski, Barbara Nichols (chief executive officer of the Commission on Graduates of Foreign Nursing Schools in Philadelphia), and Mary Powell (a postdoctoral research fellow, U.Penn.).

[Abstract by Willard Hom, Director Research & Planning, System Office California Community Colleges, completed August, 2004]

Wages: An Incomplete Measure of Labor Outcomes

Do non-wage factors influence job choice? Two economists say “yes” after they analyzed nine years of data from the Panel Study of Income Dynamics, a national survey that began tracking individuals in 1968. For their study, they used only 2,313 people from this survey; this represented those people who made voluntary job changes in the years 1984-1992. Among their findings were the following:

1. “Though the majority, 53%, move to higher paying jobs, a very large proportion of voluntary job changes, 42.5%, move to jobs that pay lower wages...The median percentage change in real wages [that is, CPI-adjusted] for those moving to lower-paying jobs is –17.8%...” [pp.4-5, 2003]
   
   
2. “A wage decline may represent the fact that the amenities of the new job are greater for the worker than those of the old job...[This] has important implications for assessing whether people’s quality of life has improved over time. Wages are often used as a measure of an individual’s standard of living. But now we see why this might be misleading...” [p.1, 2004]
   
   
3. “Job amenities are defined as various non-wage characteristics of a job, such as job stress, the general work environment, inflexible scheduling for hours or days of work, location of the employer, and so on...” [p.3, 2004] [Note: The economists also note that the potential for higher future pay can motivate the switch to a job that offers a lower starting wage. See p.5, 2003]
   
   
4. “The success of the dynamic model of job choice [constructed by these economists to explain the survey results]...tell us that job amenities do figure prominently when people decide where they want to work. An important implication of this finding is that using income alone to measure standards of living can be misleading..” [pp.3-4, 2004]

These findings should have implications for analysts of workforce development. It seems possible that the quantitative evaluation of a program may fail to capture the full improvement in worker welfare if the evaluation relies solely upon changes in reported wages of program participants (as well as wages for any comparison group used). In the past, economists have voiced this theory about non-wage factors in worker welfare, but now there is empirical evidence to support the theory. Consideration of this theory should motivate planners of evaluations to use additional outcome criteria, where possible, to avoid this limitation of wage increases as the sole program outcome.

The study’s two economists were Ed Nosal and Peter Rupert, both of the Federal Reserve Bank of Cleveland. Rupert presented a “reader-friendly” summary of their analysis in the April 15, 2004 issue of Economic Commentary (published by the Federal Reserve Bank of Cleveland and downloadable at http://www.clevelandfed.org/Research/Com2004/04-15.pdf). For the very technically inclined reader, the duo presented their study’s theoretical model, data, methods, and implications in a May 2003 working paper (“How Amenities Affect Job and Wage Choices Over the Life Cycle”), which is available from the Federal Reserve Bank of Cleveland as Working Paper 03/02.

[Abstract by Willard Hom, Director Research & Planning, System Office California Community Colleges, completed 7/23/04]

About Types of Rapid County Population Growth

Two geographers, Sutton & Day, recently published a study that identifies distinct types of U.S. counties that exhibit rapid population growth. Specifically, they found 299 counties with growth at or above 70% during the 1970-1990 span. For educational researchers and planners, such information may help in strategic planning efforts.

Sutton & Day defined rapid growth counties as those counties with population growth rates equal or greater than one standard deviation above the mean growth from 1970 to 1990. Then, using a statistical method known as k-means cluster analysis, they made the following conclusions:

1. “The fifteen variables [chosen to explain population growth] produced 7 clusters of high growth countries: (1) small urban, (2) large urban, (3) white-collar suburban, (4) blue-collar suburban, (5) retirement, (6) isolated growth/recreation, and (7) Hispanic...Four of the seven clusters were identified as urban or suburban and clearly fit the broad category of metropolitan. The remaining three clusters (retirement, isolated growth/recreation, and Hispanic) tended to be non-metropolitan in character...” [p.259]
   
   
2. [For the large urban cluster], “...Only six counties grouped in this cluster, the fewest in any cluster in the study. These six counties, however, represent truly amazing growth in terms of raw population numbers...Growth of the magnitude evident in San Diego is rare in counties which already have large populations, but nevertheless five other counties (Maricopa, AZ; Pima, AZ; Riverside, CA; San Bernardino, CA; and Broward, FL) met and even exceeded the rates of growth observed in San Diego...” [p.261]

This study, though limited to data before the 2000 census, may help policy makers to recognize (a) specific places of growth and (b) how any one of these may differ in terms of growth factors when compared to other counties with apparently equivalent growth rates. In terms of higher education’s regional planning for enrollment capacity and for workforce development, such information may be useful.

Paul D. Sutton and Frederick A. Day (both with the Dept. of Geography at Texas State University, San Marcos) presented the study’s results, background, and methods in their article (“Types of Rapidly Growing Counties of the U.S., 1970-1990”) in the refereed journal, The Social Science Journal, (2004, Volume 41, pages 251-265). The article includes their definition of the 15 clustering variables (which theoretically represented the explanatory growth drivers of either a social/demographic factor, a geographic factor, an economic factor, or a political factor) and a table to help characterize each cluster (via a tabulation of the central values by cluster for each of the 15 variables) .

[Abstract done by Willard Hom, Director, Research & Planning Unit, System Office, California Community Colleges, 8/18/04]

A Different Approach for Analyzing Transfers from 2-Year Colleges

Although public two-year colleges (or community colleges—CCs) generally have multiple missions, there has been historical interest in measuring how well they perform one of those missions, the transferring of students to four-year institutions in order to complete baccalaureate degrees. This area of research recently received an in-depth examination from two economists. Ronald G. Ehrenberg (Cornell University, School of Industrial and Labor Relations) & Christopher L. Smith (of Massachusetts Institute of Technology, Department of Economics)—or E&S—completed an extensive analysis involving data on 13,383 students who transferred from New York State’s public two-year colleges to its public four-year colleges. Their goal was to “describe a methodology that can be used to evaluate how well each 4-year public institution is performing in educating students who transfer to it from each 2-year institution in the state and how well each 2-year public institution in the state is performing in preparing its students who transfer to 4-year public institutions in the state to complete their programs at the four-year institutions....” [pp.25-6] The following are some salient points in their article:

1. “...it is important to control for student preparation and background in any attempt to evaluate the relative performance of 4-year institutions within a state system in educating transfer students...” [p.24] [Note: Student-level data that includes levels of achievement and resources at different points in time are critical.]
   
   
2. [to control for differences in credits earned by students who transfer to different 4-year colleges] “...include information on the number of credits toward the 4-year degree that each student in our sample received at the time of enrollment...However, in the absence of the availability of such data, a simpler approach is simply to eliminate from the sample all of the individuals who transferred without first receiving a 2-year college degree specifically designed to prepare them for transfer to 4-year colleges...” [p.20] [Note: In testing the latter option, E&S found it shrank their sample to 5,271 students---or more than a 50% reduction in usable cases]
   
   
3. “...once we control for the 6-year graduation rate of freshmen, the ranking of the 4-year institutions on the probability that transfer students to them graduate within three years is highly positively correlated with the share of an institution’s new students that are transfer students...Four-year institutions that accept large shares of their student bodies via the transfer route appear to become especially adept at assuring that transfer students make normal progress...” [pp.24-5] [Note: If this phenomenon is generalizable to other states, it may help CC transfer staff and CC students in decisions about 4-year college destinations.]
   
   
4. “...Accountability rankings of this type are best thought of as formative rather than summative. Rather than using them to reward, or penalize, institutions, it would be more productive, at least in the initial years that they are employed, for system administrators and policy makers to focus on outliers to try to learn what the factors are that cause some 4- and 2-year institutions to ‘look better’ or ‘look worse’ on these measures than do other institutions within the system. Once the factors are discovered, to the extent that they relate to practices of institutions, rather than characteristics of the institutions, their students, or the areas in which they are located that are not accounted for in the estimation, dissemination of information about the actions taken by the ‘better institutions’ that led to their success to all institutions in the system would be beneficial...” [p.26] [Note: E&S withheld any information that could lead to a public ranking of the institutions in the study. This helped maintain focus on variables that policy makers can control to some extent, and it prevented the potentially distracting misuse of results to criticize specific school administrators for outcomes beyond their control.]
   
   
5. “Rather than assuming, as we have done..., that the admission standards and preparation of transfer students to each 4-year institution can be controlled for by the 6-year graduation rate of students who initially enroll as freshmen at the institution, it would be by far preferable to try to directly control for these variables. If data on academic backgrounds and information on each transfer student are not easily available...,information on minimum grade point average, or the average grade point average, for transfers that each 4-year institution has admitted in each year would be useful...” [p.26] [Note: This primarily applies to analyses of 4-year institutional performance.]
   
   
6. “...Having detailed data on the nature of articulation agreements between each 2-year campus and each 4-year campus within a state system, whether each agreement is adhered to and the resources that each institution applies to advising potential transfer students from or to it would improve the analyses and aid in the interpretation of the estimated institutional coefficients...” [p.26] [Note: Measuring this factor will challenge any researcher because of the great variability in it---which also ironically contributes to its importance as a potential, explanatory variable.]
   
   
7. “...if a system has the resources, staff talent and time to estimate models [a superior approach is advised]...This type of approach would have taken account of the fact that individual-level and institutional-level data...would be included as explanatory variables. One, but not the only, way to approach such data is to use hierarchical linear models (HLM)...” [p.27]
   
   
8. “...more precise estimates could be obtained using the method of weighted regression. In addition, the linear probability model does not take into account that each of the probabilities can vary only between 0 and 1 [true of variables expressed as probabilities, proportions, or percentages], and that there is an explicit ordering of the probabilities (graduating is better than still being enrolled, which in turn is better than having dropped out). Using a multinomial logit model takes the first into account, while using either an ordered probit or logit model model takes the second into account...[p.26] [Note: E&S report that results with these more precise methods were still similar to those using ordinary least squares regression. The advanced methods mainly add precision.]

This article has relevance for both researchers and administrators in public higher education. For researchers, it presents some ways to conduct an evaluation of transfer outcomes at the four-year level and at the two-year level. It clearly indicates the need for a vast set of data and extensive staff analytical expertise. For administrators, the relevance may lie in the conceptualization of transfer outcomes as a process that embraces the four-year level, as well as the two-year level. To limit the analysis of transfer processes to the matriculation of a two-year student at a four-year college seems to omit a substantial part of the benefit that the public expects to obtain from a public higher education system. Finally, the study demonstrates how a study of institutions can proceed to inform administrators for process improvement without the hazards of publishing popular, so-called rankings.

E&S document their study’s theoretical background, data, methods, limitations, and implications in their article (“Analyzing the Success of Student Transitions from 2- to 4-Year Institutions within a State”) published on pages 11-28 (2004, Vol.58, No.5) of the journal Economics of Education Review. Incidentally, E&S wrote a preliminary draft report issued in February 2002, and it is available at the following site: http://www.ilr.cornell.edu/cheri/wp/cheri_wp22.pdf

[Abstract by Willard Hom, Director Research & Planning, System Office California Community Colleges, completed 7/19/04]

On Reporting the Effect Size of a Study

Quantitative researchers use the term effect size to conceptualize and express the strength of a statistical relationship in their analyses. The effect size is critical for interpreting study results because the traditional reporting of tests of statistical significance does not provide information about strength of relationship. An article in the Fall 2003 issue of The Journal of Experimental Education (Vol. 72, No.1, 51-64) discusses the issue of reporting effect sizes for studies. The authors stated the following:

1. “In three experiments, we examined the influence of effect size information (Cohen’s d) and commentary about effect sizes on readers’ perceptions of research results. Both conditions were examined by having undergraduate students, graduate students, and faculty members read an authentic journal article in which effect size information, commentary about effect size information, or nothing was included in addition to descriptive and inferential statistical information...” [p.59]
   
   
2. “...it is the commentary about effect size, more than the inclusion of the effect size, that affects undergraduate readers. When a comment was included, undergraduates rated the results as being more important than when commentary about effect size was not included, regardless of whether Cohen’s d was included. In contrast, graduate students’ and faculty members’ importance ratings were not affected by the inclusion of commentary concerning effect size information...” [p. 59]
   
   
3. “Wainer & Robinson (2003) perused recent issues of educational psychology journals to determine how authors presented effect size information and how they commented on those effects. Sadly, Wainer & Robinson found that most authors seem to follow arbitrary criteria in categorizing effect sizes [i.e., the .2/.5/.8 classification]—a practice that even Cohen himself warned against...” [pp. 59-60]
   
   
4. “Knapp & Sawilowsky (2001)...recommended that people with content expertise should decide whether results are practical and meaningful. Perhaps editorial policies should state that authors who lack such expertise should simply present their findings without such arbitrary commentary as ‘small,’ ‘moderate,’ and ‘large.’...[p.60]
   
   
5. “...it appears that the effect of including commentary about effect size information may depend on the reader. Graduate students and faculty in educational psychology are typically more proficient in statistics than are undergraduate students, which may explain a greater reliance that the undergraduate students placed on the commentary concerning the effect size information...Rather than require authors to include information that may result in misleading the untrained reader, perhaps editors and reviewers should carefully consider which effect sizes to report, if any, and how they might be interpreted by readers...When considering effect size and commentary information, perhaps it would be best to simply provide either sufficient statistics (so that an effect size can be calculated) or an effect size (if it can be interpreted meaningfully), and let a content expert decide whether the results are important or impressive in a practical context. This would solve the problem of interpreting the size of the effect for readers...” [p. 60]

The authors of the article were Daniel H. Robinson (U.Texas); Tiffany A. Whittaker (U. Missouri-Columbia); Natasha J. Williams (American College Testing); and S. Natasha Beretvas (U. Texas). The article indicates that, despite a widespread need to consider effect sizes in research, the way researchers report effect size can possibly mislead or influence the perceptions of readers, particularly readers with less proficiency in statistical analysis.

[Abstract by Willard C. Hom, Director, Research & Planning Unit, Chancellor’s Office, California Community Colleges, 5/17/04]

New Evidence on the “No Opinion” Response Option

Jon A.Krosnick (a professor of psychology and political science at Ohio State University) and a team of survey researchers recently published the results of an analysis that should help survey researchers and designers to understand the dynamics of the “no opinion” or “don’t know” response. Over the past few decades, survey researchers have followed the advice of some pioneers in surveys by including either a “no opinion” or a “don’t know” option in their attitude questions. This tradition presumably improved the quality of the attitude measures because it reduced the tendency for respondents to give “incorrect” answers when they lacked an opinion or enough knowledge to answer the questions posed.

However, the new findings are as follows:

“…Inclusion of a no-opinion option did not reliably improve the quality of the data obtained…These findings therefore question the notion that no-opinion options discourage respondents from providing meaningless answers to survey questions. Instead, this evidence is consistent with the possibility that the responses attracted by no-opinion options would have provided substantive answers of the same reliability and validity as were provided by people not attracted to those options…Attraction to no-opinion response options was most common among respondents of low education, when respondents voted secretly, when questions were asked late in a survey, and when respondents devoted little effort to answering questions…As appealing as offering a no-opinion response may be, doing so may lead researchers to collect less valid and informative data than could be done by omitting it…But, in fact, offering this option will most likely reduce a researcher’s effective sample size (by encouraging some respondents to say “don’t know”), yielding reduced statistical power at the very least. Furthermore, a no-opinion option apparently systematically encourages low-education respondents to avoid the effort of deciding how to answer the question, thereby reducing the impact they have on survey results. If researchers want to collect as many valid opinions as exist, it appears that doing so may best be done by omitting no-opinion options and measuring attitude strength directly instead.”

This study has relevance for institutional researchers. They frequently design, administer, and analyze surveys, and the issue of using an option for “no-opinion” or “don’t know” has periodically caused concern and debate. This study used experiments within some large-scale surveys over a period of years and across diverse topics, and its persuasive findings will probably influence the next generation of survey designs.

Interested parties may find the Krosnick, et al, article, “The Impact of ‘No Opinion Response Options on Data Quality’,” in the following journal: Public Opinion Quarterly, Volume 66, Number 3, Fall 2002, 371-403.

[Abstract by Willard Hom, Director Research & Planning, Chancellor’s Office California Community Colleges]

Links to Other Publications of Interest

This section lists some web pages that may interest community college staff. The text descriptions are basically excerpts from the announcements appearing on listserves. As usual, some web addresses may change over time, and you may occasionally encounter an inoperative link (in which case we apologize in advance).

Student Learning/Participation

Practical Assessment, Research & Evaluation, Vol 9, No. 16; "Course Selection Decisions by Students on Campuses With and Without Published Teaching Evaluations"; Wendy Bryce Wilhelm & Charles Comegys (2004).
http://PAREonline.net/getvn.asp?v=9&n=16

Center for Studies in Higher Education (CSHE) Report; "Learning and Academic Engagement in the Multiversity--Results of the First University of California Undergraduate Experience Survey;" Richard Flacks, Gregg Thomson, John Douglass & Kyra Caspary; June 2004. (noted by Patrick Perry)
http://ishi.lib.berkeley.edu/cshe/seru21/

Pew Hispanic Center Report; "Latino Youth Finishing College: The Role of Selective Pathways"; Richard Fry; June 23, 2004. (noted by Linda Umbdenstock)
http://www.pewtrusts.com/pdf/pew_hispanic_college062304.pdf

Teaching & Faculty

NCES Report; "Teacher Attrition and Mobility: Results from the Teacher Follow-up Survey, 2000-01"; Kathy Chandler; August 2004. This new report provides information about teacher mobility and attrition. It is a one-year follow-up of a sample of teachers who were originally selected for the Teacher questionnaire of the Schools and Staffing Survey.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2004301

NCES Issue Brief; "Who Teaches Reading in Public Elementary Schools?"; David Meyer & Daniel McGrath; August 2004. This new Issue Brief describes the qualifications of public school elementary-level reading teachers relative to general elementary teachers, in terms of educational attainment, educational preparation and certification in reading, and educational preparation and certification in elementary education.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2004034

Workforce Development

U.S. Census Bureau Report; "Dynamics of Economic Well-Being: Movements in the U.S. Income Distribution, 1996-1999"; John J. Hisnanick & Katherine G. Walker; July 2004. (noted by Steve Maack)
http://www.census.gov/prod/2004pubs/p70-95.pdf

CCC Chancellor’s Advisory Committee on Work-Based Learning and Employment Services' Study; "The Benefits of Work-Based Learning and Occupational Coursework in the California Community Colleges"; Svetlana Darche, Michael P. Arnold, and Corey Newhouse; May 3, 2004. A study of community college students and their employment outcomes (noted by Chuck Wiseley).
http://www.calworkplace.org/WBL%20Report.pdf

Strategic Planning

NCES Report; "Federal Support for Education: Fiscal Years 1980 to 2003"; William C. Sonnenberg; August 2004.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2004026

Education Statistics Quarterly; Vol. 5, Issue 4; August 2004. This issue includes coverage of IT issues, among other things.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2004610

California Performance Review (CPR) Report. To review the CPR report in its entirety, see the link below. It includes actions regarding community colleges.
www.report.cpr.ca.gov

NCES Report; "Language Minorities and Their Educational and Labor Market Indicators—Recent Trends"; Steven Klein, Rosio Bugarin, Renee Beltranena, and Edith McArthur; June 2004. This report examines trends in the characteristics of the U.S. language minority population from 1979 through 1999.
http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2004009

PEW Internet & American Life Project Report; "The Internet and Daily Life"; Deborah Fallows; August 11, 2004. A Pew report on daily Internet use by the public.
http://www.pewtrusts.com/pdf/pew_internet_081204.pdf

PPIC Report; "PPIC Statewide Survey"; Mark Baldassare; August 2004. An objective, in-depth profile of the social, economic, and political forces affecting public policy preferences and ballot choices in California.
http://www.ppic.org/content/pubs/S_804MBCAGS.pdf

Research Methods

NAAL Survey; "2003 National Assessments of Adult Literacy (NAAL): English Background Questionnaire"; 2003. This survey collects data on a variety of background variables, many of which obtain valuable information not collected in the 1992 survey.
http://nces.ed.gov/naal/design/design.asp#instrument

PPIC California Counts; Vol. 6, No. 1; "California’s Multiracial Population"; Laura E. Hill, Hans P. Johnson, and Sonya M. Tafoya; August 2004. This issue covers the size and nature of the state’s multiracial population and implications for researchers.
http://www.ppic.org/content/pubs/CC_804LHCC.pdf