The structure of school career development interventions: implications for school counselors. (Special issue: career development and the changing workplace).

Note: The work reported herein was supported under the National Dissemination for Career and Technical Education, PR/Award (No. VO51A 990004) as administered by the Office of Vocational and Adult Education, U.S. Department of Education. However the contents do not necessarily represent the positions or policies of the Office of Vocational and Adult Education or the U.S. Department of Education, and the reader should not assume endorsement by the federal government.

Each year, large sums of state and federal dollars are spent on career development provided through America's public schools. The annual federal expenditure alone on career development interventions exceeds one billion dollars (Carl D. Perkins Vocational and Technical Education Act, 1998). Such expenditures are seen as a sound investment, for it is believed that they increase the relevance of education for America's youth (Herr & Cramer, 1996). However, despite these benefits, researchers have not attempted to define the actual structure of career development interventions that occur in America's secondary schools. Without the knowledge of such a structure, real progress in career development interventions research and practice will be either severely circumscribed or absent.

This taxonomic research project was undertaken in order to develop a taxonomy of career development interventions provided through America's secondary schools. In his monograph on typologies and taxonomies in the social sciences, Bailey (1994) commented that "without classification, there could be no advanced conceptualization, reasoning, language, data analysis, or for that matter social science research" (p. 1).

The production of a career development interventions taxonomy has benefits for both researchers and practitioners. First of all, a taxonomy would bring a standardized nomenclature both in terms of content and structure to this area of classifying interventions. With this standardization, different initiatives then can be accurately compared and contrasted (Brickell et al., 1974). One of the major problems noted in the reviews of early career development research was the lack of definition of the treatment. Because of this lack of definition, the reviewers found it difficult to draw an accurate conclusion about the relationship between career development interventions and important outcomes such as student achievement (Bhaerman, 1977; Bonnet, 1977; Evans & Burck, 1992; Hoyt, 1980).

When a field lacks a taxonomy, it is difficult to compare the efficacy of certain types of interventions against other sets of interventions (e.g., guidance curriculum vs. experience-based learning). This lack of comparative research leaves school counseling practitioners with no alternative other than employing an inefficient and expensive "shotgun" approach to career development for America's youth.

Brickell et al. (1974) held that since taxonomies can be used to generate a set of distinctive choices, they can help frame the debate about what choices to make. Thus, policymakers at all levels can draw useful guidance from them. With a taxonomy, school counseling researchers can explore systematically and comprehensively questions about the relationship between career development interventions and desired student outcomes. For example, one outcome question might be: "Does the sequencing of career development interventions taxa interact with a successful transition to postsecondary education?" Another important question that the production of a taxonomy would open for study is: "For students of color, which career development intervention taxa most leverage academic achievement?"

A taxonomy can also help with other formative evaluation questions. For example, a taxonomy of career development interventions would give a school counselor a parsimonious framework within which to examine if there are areas of career development where her or his district is underdeveloped (Bailey, 1994). Moreover, the school counselor could use this taxonomy to map programmatic progress (Brickell et al., 1974). For example, using a taxonomy as a guide, a group of school counselors discovers that their district does a lot of career-related curriculum activities but no activities at work sites in the community. Thus, these school counselors could become advocates in their district for a more balanced career development program.

METHOD

Participants

The participants for this study were drawn from a random sample of 12% (n = 194) of the membership of the Guidance Division of the Association for Career and Technical Education (ACTE). The following were excluded from the sample: (a) members of the research team (n = 1), (b) surveys returned by the postal service with no forwarding address (n = 7), and (c) surveys returned by potential respondents claiming lack of ability or desire to complete the instrument (n = 6). Thus, the final sample size was 179. Of the 179, a total of 101 surveys were returned for a 60% return rate. This return rate is a strong one given the length of the survey which contained 220 items. The participants in this study were highly educated and had extensive professional experience in the career development field. The vast majority (88%) possessed at least a master's degree. The average number of years of professional experience with career development activities was 15. The majority were female (71%); persons of color represented 16% of respondents. A little more than three quarters of the participants worked in K-12 settings (76%). Approximately one half of the participants worked as school counselors (55%). The professions of the other respondents were teacher (13%), administrator (14%), paraprofessional (1%), and other (13%).

Instrument

Development of comprehensive interventions list. We examined research articles, grant reports, and program manuals in order to develop a comprehensive list of career development interventions that commonly occur in American secondary schools. Drafts of this list were circulated to career development professors and practitioners around the country for feedback. These career development professionals were asked to recommend additions and subtractions to the list as well any nomenclature revisions. The research team incorporated their feedback into a final list of 44 interventions. This list represented the first comprehensive articulation and demarcation of the universe of career development interventions. This list formed the foundation for the survey sent to the participants.

Determination of rating variables. Developing a comprehensive listing of interventions is only the first step in the creation of a taxonomy of interventions. The next key step is selecting the variables upon which to cluster the interventions. A total of 13 variables that can be used to describe school-sponsored programming were considered.

If all 13 variables were used to rate the 44 interventions, the subsequent survey would have contained 572 items. In his classic text on mail survey research, Dillman (1978) noted that response rate reductions can be expected in surveys that exceed 125 items. Thus, we decided to limit the number and cognitive complexity of the ratings in the hope of securing an acceptable return rate. The limits we set were: (a) five variables times 44 interventions in terms of length, and (b) binomial ratings in terms of cognitive complexity. These limits resulted in a survey with a total number of items slightly under twice the number of Dillman's reduced response-rate cut point.

Once a limit of five variables was set, the question of which five variables to use emerged. Fortunately, there exists a quality continuum for variables used to perform a cluster analysis. On one end of the continuum there are the most "robust" variables and on the other end there are the most "noisy" variables. By robust we mean variables that lead to the most homogeneous clusters possible (Carmone, Kara, & Maxwell, 1999). By noisy we mean variables that mask the clear structure portrayed by the other variables and yield less homogeneous clusters (Carmone et al.).

To determine which 5 variables to employ, we first placed all 13 variables along the quality continuum described in the previous paragraph. This placement was done conceptually and was guided by Saylor and Alexander's (1966) influential text on curriculum planning. In particular, we were influenced by their six-level model of factors that influence "the actual experiences provided pupils by the school" (p. 273). We rated as most robust those variables that occurred on the factor level that Saylor and Alexander held was most closely linked to actual student experiences. The five variables placed closest to the Robust pole were selected for use in the study. The selected variables were: (a) duration of the intervention [Time], (b) pedagogical style [Mode], (c) location of intervention [Place], (d) size of the intervention group [Size], and (e) control of the intervention's application [Control].

Intervention rating procedures. On the survey itself, participants were asked to select the binomial descriptor that "best" represented an intervention. Brief definitions were provided for each descriptor. Specifically, for the Time variable, the participant could circle S for short (1 hour to 2 weeks) or L for long (greater then 2 weeks). For the Mode variable, the participant could circle P for passive (listening, reading, viewing) or A for active (hands on). For the Place variable, the participant could circle S for school based (on campus) or C for community based (off campus). For the Size variable, the participant could circle G for group (small group and above) or I for individual (a student). Finally, for the Control variable, the participant could circle Y for youth directed (control over time/scope/location) or A for adult directed (control over time/scope/ location). In total, each participant made 220 ratings.

Data Analysis

Cluster analysis. Bailey (1994) noted that the most common technique to establish taxonomies in the social sciences is cluster analysis. In their article on the uses of cluster analysis in counseling psychology research, Borgen and Barnett (1987) noted that clustering could be an especially powerful technique for simplifying complex data sets. They stated "it can be used to group the objects when use of human judgment would be tedious, subjective, or practically impossible" (p. 461). At its most basic level, cluster analysis is the application of a set of rules (i.e., algorithm) for dividing a proximity matrix in order to form groups of similar objects (Borgen & Barnett). In this definition, the term proximity matrix refers to a matrix of similarities or dissimilarities between objects. The cluster analysis conducted in this study was accomplished by means of the ClustanGraphics software program (Wishart, 1999), a program that is highly regarded in the taxonomy research community (Bailey).

When cluster analysis is performed using ratings from scales utilizing different metrics, raw scale scores are converted to a standard score. In the present study, such a transformation was not performed because the five variable ratings all used the same metric. Concerning the particular data set of this study, Wishart (personal communication, May 9, 2000) recommended that cluster analysis be conducted by first computing proximities via the Euclidean Sum of Squares method. Then he recommended that the proximities be clustered by the Increase Sum of Squares method. These recommendations were followed exactly. In addition to Wishart's recommendation, there is a large body of literature that reports the Increase Sum of Squares method as the preferred method of clustering (e.g., Borgen & Barnett, 1987; Milligan & Cooper, 1987).

To answer the question of how many taxons are present in the data, we employed the best cut procedure available in the ClustanGraphics program. This procedure uses significance tests based on the fusion values. We used the default Upper Tail test of significance set at the .05 level. Besides clustering, the ClustanGraphics program also notes the case (i.e., career development intervention) that statistically most exemplifies each taxon. Taxon exemplars are the cases which have the minimum within-cluster average dissimilarity.

Modal rank score computation. In addition to the cluster analysis, the modal quintile rank score for each variable on each taxon was computed. By the term modal we mean the most frequent response (Vogt, 1993). In order to facilitate taxon to taxon comparisons, descriptors were assigned to each quintile rank score on each variable. For instance, on the Time variable the descriptor assigned to the first 1st quintile was Strong Short. The descriptors for the other quintiles of the Time variable were Short (2nd quintile), Mixed (3rd quintile), Long (4th quintile), and Strong Long (5th quintile).

RESULTS

ClustanGraphics' best cut procedure reported a significant four taxon solution. This four taxon solution was significant at .05 on an Upper Tail Test with 43 degrees of freedom (Realized Deviates = 1.54, t-statistic = 10.07). As with factor analysis, naming taxa can be as much an art as it is a science. Many names were considered for each taxon. In the end, we chose to apply nomenclature that we believe reflected the taxon exemplar. The first taxon's exemplar was Youth Apprenticeship. This taxon contained 10 interventions and these interventions occur at work sites in community settings rather than school settings. Thus, we choose the taxon name Work-Based Interventions. The second taxon was named Advising Interventions by drawing upon the taxon exemplar Career Peer Advising/Tutoring. This taxon contained 20 interventions. The third taxon contained the exemplar Career Day/Career Fair. Thus, we named this taxon Introductory Interventions, and it contained seven interventions. The final taxon had Tech Prep/2+2 Curriculum as its exemplar. Thus, we labeled this taxon Curriculum-Based Interventions given the structural and instructional nature of the exemplar. This taxon also contained seven interventions. The Table lists the 44 interventions by taxon. The member interventions for each taxon are listed alphabetically below the taxon titles. Computation of a modal quintile rank scores for each variable on each taxon was also completed. See the Figure for the results of these computations.

DISCUSSION

This study produced the first taxonomy of the career development interventions that are provided through American secondary schools. The results suggest that there is an underlying four-part structure to this universe of interventions. The four intervention taxa were denoted Work-Based Interventions, Advising Interventions, Introductory Interventions, and Curriculum-Based Interventions. In this section, we provide a detailed description of each taxon and present the potential implications of the study.

Taxon Descriptions

Work-based interventions. This taxon was defined as follows: "The class of interventions designed to promote both career and academic self-efficacy and motivation through sustained and meaningful interactions with work sites in the community." This taxon was the only one with a Community modal descriptor for the Place variable. It was also the only taxon with a Strong Active descriptor for Mode.

Advising interventions. This taxon was defined as follows: "The class of interventions designed to provide direction, resolve impediments, or sustain planfulness in students about their goals for the future." Super and Overstreet (1960) felt that planfulness was a core element to successful career development in adolescence. This taxon was the only one of the school-based taxa described as delivered on an individual basis.

Introductory interventions. This taxon was defined as follows: "The class of interventions designed to awaken a student's interest in their own personal and professional growth." This taxon was similar to the fourth taxon with one exception. The modal Time variable descriptor for this taxon was Strong Short and the modal Time descriptor for the fourth taxon was Strong Long. This temporal difference emphasizes the introductory and preparatory nature of this class of interventions.

Curriculum-based interventions. This taxon was defined as follows: "The class of interventions designed to promote career and academic knowledge and skills through means and content relevant to the world of work." As noted in the previous section, the distinguishing characteristic of this taxon of interventions versus the other school-based taxa is its long-term nature. The interventions contained in this taxon represent the school-based complements to the interventions that are part of the Work-based Interventions taxon.

Implications

We felt that four implications for school counselors and researchers emerged from this taxonomic study. These implications were (a) improved communication among school counselors, (b) improved formative evaluation of career development interventions, (c) improved summative evaluation of career development interventions, and (d) improved research of career development interventions.

Improved communication. A uniform nomenclature and definition are critical to adequate communication among all professionals. Without such a nomenclature and definition, misunderstanding and confusion are commonplace. For example, a school district guidance coordinator suggests that a school district institute a practicum program. As a result, one of the counselors in the district does some research online and finds multiple definitions of this term. Also, some of definitions sound the same as the work-based learning project program and the internship program the district already has in place. Given this blurring of program titles and activities, the school counselor becomes confused about what to do next and thus decides to set aside the guidance coordinator's suggestions. Another example of confusion concerns the school counselor who goes to a state convention and learns about the advantages of starting a career majors program. The school counselor wonders how this program is similar to the career pathways program the school counselor read about in a recent publication. The school counselor is unsure whom to follow and finds his or her enthusiasm wanes.

The taxonomy produced by this study may also help school counselors engage in richer program planning. The number of distinct interventions we discovered in our research exceeded 40. A number this high presents a level of complexity that inhibits reflection on, and organization of, professional practice. By their very nature, taxonomies accurately "chunk" large sets of interventions. Thus, with our taxonomy, school counselors planning a state convention do not have to wade through an endless set of disconnected interventions in order to develop a convention theme and program. Instead, they can select to focus on building capacity in a certain role in a certain area. (e.g., building consulting capacity with the interventions of the curriculum taxon.)

Improved formative evaluation. By the term formative evaluation we mean evaluation undertaken to find ways to improve, redesign, or fine-tune a program (Vogt, 1993). As noted earlier, the universe of career development interventions is presently disconnected and ill-defined. AS such, it is difficult for school district personnel to answer comprehensively and in detail questions such as "In terms of career development interventions, where is our district over-programmed and where is district under-programmed?" However, with the taxonomy reported in this study, along with its associated nomenclature and definitions, a school counselor gains a language with which to ask answerable questions. The following question is an example: "Compared to other local schools, is my school under-programmed in reference to work-based interventions?"

In addition to creating the space for intelligible programming questions, the creation of this taxonomy gives school counselors the tools to evaluate whether gaps exist between theory and practice. For instance, the placement of the guidance lesson interventions in the Introductory Interventions taxon was unexpected in terms of present school counseling theory. Influential texts on school counseling (e.g., Campbell & Dahir, 1997; Gysbers & Henderson, 2000) recommend that these interventions be seamlessly woven into the curriculum and be long term. Thus, on a theoretical level, the three guidance lesson interventions should have been found in the fourth taxon. Instead, the random national sample of guidance professionals used in this study reported viewing these interventions' as short term. This description suggests that in practice, guidance lessons are still delivered by school counselors in an intermittent and time-limited fashion. Thus, the results of this taxonomic study suggest critical gaps may exist between best practice and actual practice in the career development arena.

Improved summative evaluation. By the term summative evaluation we mean evaluation undertaken to assess the impact of a program or to determine how well a program has met its goals (Vogt, 1993). A full description of program components is foundational to quality summative evaluation (Wholey, Harry, & Newcomer, 1994). Unfortunately, such description has not been the tradition in the summative evaluation of career development intervention programming. For example, reviewers of the Career Education (CE) program evaluation literature noted there were many methodological problems (Bhaerman, 1977; Bonnet, 1977; Datta, 1977; Hoyt, 1980). One of the key problems cited was the poor definition of the "treatment" in these evaluations. Concerning CE as treatment, Bhaerman commented that:

   As with the length of the data collection period,
   it is difficult to generalize since, as stated
   before, it was not always possible to ascertain-from
   the printed reports--the degree of
   the "delivery of program components." (One
   researcher, M. F. Smith, spoke of this as "the
   nebulous variable"--and it was). I wish it were
   possible to illustrate clearly and simply the
   direct relationship between scope and intensity
   of the CE treatment and the results. But,
   because of the nature of the reports, a one-to-one
   relationship cannot be easily drawn. (p. 52)

In the end, the shortcomings of the CE summative evaluations were the result of the investigators' failure to adequately define the treatment. However, poor summative evaluation execution on this question was not limited to CE evaluators.

The only large-scale, post-CE summative evaluation concerning the impact of career development interventions on academic achievement was Lapan, Gysbers, and Sun (1997). Lapan et al. found that academic achievement (as measured by self-reported grades) was positively related to level of guidance program implementation. However, this study contained the conceptual and methodological problems present in the CE studies. Namely, that the "treatment" in this research lacked the needed specificity. In the case of Lapan et al., the sole independent variable was the Missouri Comprehensive Guidance Program in toto. However, in reality, this variable is multifaceted and contains many non-career development interventions as well.

The taxonomy presented in this study fills the treatment specificity gap noted by Bhaerman (1977). This taxonomy and its associated nomenclature and definitions provide school counselors the ability to conduct comprehensive evaluations of their career development intervention work. The following is an example of an evaluation question: "Did the curriculum taxon interventions increase student interest in taking more advanced classes in math and science?"

Improved research. By the term research we mean the systematic investigation of a subject aimed at uncovering new information (discovering data) and/or interpreting relations among a subject's parts (theorizing; Vogt, 1993). Research is similar to summative evaluation in that they both focus on outcomes. However, research differs from program evaluation in that the end of research is generalizable rather than local knowledge. In terms of career development intervention research, the same problem present in summative evaluation is also present here. Namely, the problem of ill-defined treatments. Thus, it is difficult to conclude from past research on school-based career development interventions why certain outcomes occurred. Prominent reviewers of career development intervention research have noted this problem. For example, Whiston and Sexton (1998) conducted a major review of the school counseling outcome literature published from 1988 to 1995. In this review, they noted how research methodology problems made it difficult to draw any definite conclusions about the relationship between guidance and academic achievement.

As with summative evaluation, the taxonomy presented in this study fills the treatment specificity gap. This taxonomy and its associated nomenclature and definitions give school counseling researchers the foundational tools needed to study causal relationships between interventions and outcomes. The following is an example of a research question: "What dosage of introductory interventions is sufficient to raise student career maturity?"

CONCLUSION

In the social sciences, taxonomic studies are rarely conducted as ends in themselves. These studies are completed in order to provide tools for researchers and practitioners. The lack of specificity about interventions in the career development and academic achievement literature noted in this article is a problem for school counselors and policymakers in an era of limited resources. The nation's school counselors need concrete guidance on what career development activities can give them the most leverage in promoting student success. The taxonomy described in this report brings needed clarity, specificity, and organization to the treatment side of career development intervention research projects. In addition, this taxonomy enhances the school counselor's ability to comprehensively and systematically evaluate and monitor their career development efforts.

Figure. Modal Quintile Rank Descriptors by Taxon and Variable

Taxon                               Variable

                     Time             Mode          Place

Work based           Long        Strong Active    Community

Advising School       *              Mixed          Strong

Introductory      Strong Short      Active       Strong School

Curriculum        Strong Long       Active       Strong School

Taxon                     Variable

                      Size       Control

Work based           Strong       Mixed
                   Individual
Advising School    Individual     Adult

Introductory      Strong Group    Adult

Curriculum        Strong Group    Adult

Note. Asterisk designates a tie between the modal
categories of Short and Mixed.

Interventions by Taxon

Work-based                Advising
Interventions             Interventions

Cooperative Education     Academic Planning Counseling

Internship                Career Focused Parent/Student
                            Conference

Job Shadowing             Career Peer Advising/Tutoring *

Job Coaching              Career Map

Job Placement             Career Maturity Assessment
                          Career Counseling

Mentorship Programs       Career Interests Assessment

Service Learning/         Career Library/Career Resource
  Volunteer Programs        Center

Work Based Learning       Career Cluster/Pathway/Major
  Project                 Career Passport/Skill Certificate

Work Study                College Admissions Testing

Youth Apprenticeships *   Computer Assisted Career
                            Guidance
                          Cooperative/Dual Enrollment
                          Information Interviewing
                          Job Hunting Preparation
                          Personal/Social Counseling
                          Portfolio/Individual Career Plan
                          Recruiting
                          Referral to External Training
                            Programs
                          Referral to External
                            Counseling/Assessment

Introductory           Curriculum-based
Interventions          Interventions

Career Day/Career      Career Information
  Fair *                 Infused Into
                         Curriculum
Career Field Trip
                       Career/Technical
Career Aptitude          Education Course
  Assessment
                       Career Skills Infuse
Community Members        Into Curriculum
  Teach In Classroom
                       Career Academy/
Guidance Lessons on      Career Magnet
  Personal/Social        School
  Development
                       School-Based
Guidance Lessons on      Enterprise
  Career Development
                       Student Clubs/
Guidance Lessons on      Activities
  Academic Planning

                       Tech Prep/
                         2+2 Curriculum

Note: Asterisk designates taxon exemplar.

References

Bailey, K. D. (1994). Typologies and taxonomies: An introduction to classification in the social sciences. Thousand Oaks, CA: Sage.

Bhaerman, R. D. (1977). Career education and basic academic achievement: A descriptive analysis of the research. Washington, DC: U.S. Government Printing Office.

Bonnet, D. G. (1977).A synthesis of results and programmatic recommendations emerging from career education evaluations in 1975-1976. Washington, DC: U.S. Government Printing Office.

Borgen, E H., & Barnett, D. C. (1987). Applying cluster analysis in counseling psychology research. Journal of Counseling Psychology, 34, 456-468.

Brickell, H. B., Aslanian, C. B., Grossman, R., Spak, L. J., Callaway, M., & Smith, K. S. (1974). Taxonomy and profiles of career education (Vol. I). New York: Policy Studies of America, Educational Research Council of America. [ERIC Document Reproduction Service No. ED 106 516]

Campbell, C. A., & Dahir, C. A. (1997). National standards for school counseling programs. Alexandria, VA: American School Counselor Association.

Carl D. Perkins Vocational and Technical Education Act, 20 U.S.C. [section] 2301 (1998).

Carmone, F. J., Kara, A., & Maxwell, S. (1999). HINoV: A new model to improve market segment definition by identifying noisy variables. Journal of Marketing Research, 36, 501-509.

Datta, L. (1977). Career education: What proof do we have that it works? Washington, DC: Dept. of Health, Education, and Welfare, [Education Division], Office of Education.

Dillman, D. A. (1978). Mail and telephone surveys. New York: John Wiley.

Evans, J. H., & Burck, H. D. (1992). The effects of career education interventions on academic achievement: A meta-analysis. Journal of Counseling and Development, 71, 63-71.

Gysbers, N. C., & Henderson, P. (2000). Developing and managing your school guidance program (3rd ed.). Alexandria, VA: American Counseling Association.

Herr, E. L., & Cramer, S. H. (1996). Career guidance and counseling through the lifespan (5th ed.). New York: HarperCollins.

Hoyt, K. B. (1980). Evaluation of K-12 career education: A status report. Washington, DC: U.S. Government Printing Office.

Lapan, R. T., Gysbers, N., &Sun, Y. (1997). The impact of more fully implemented guidance programs on the school experiences of high school students: A statewide evaluation study. Journal of Counseling and Development, 75, 292-302.

Milligan, G. W., & Cooper, M. C. (1987). Methodology review: Clustering methods. Applied Psychological Measurement, 11, 329-354.

Saylor, J. G., & Alexander, W. M. (1966). Curriculum planning for modern schools. New York: Holt, Rinehart, & Winston.

Super, D. E., & Overstreet, P. (1960). The vocational maturity of ninth grade boys. New York: Bureau of Publications, Teachers College, Columbia University.

Vogt, P. W. (1993). Dictionary of statistics and methodology. Newbury Park, CA.

Whiston, S. C., &Sexton, T. L. (1998). A review of school counseling outcome research: Implications for practice. Journal of Counseling and Development, 76, 412-427.

Wholey, J. S., Harry, H. R, & Newcomer, K. E. (1994). Handbook of program evaluation. San Francisco: Jossey-Bass.

Wishart, D. (1999). ClustanGraphics primer: A guide to cluster analysis. [software] Edinburgh, Scotland: Clustan Ltd.

Cass Dykeman, Ph.D., is an associate professor, Oregon State University, Corvallis. E-mail: [e-mail]

Chris Wood, Ph.D., is an assistant professor, Department of Educational Psychology, University of Arizona, Tucson.

Michael A. Ingram, Ed.D., and Dale Pehrsson, Ed.D., are assistant professors, and Naomi Mandsager, Ph.D., is an adjunct professor; all are with Oregon State University, Corvallis.

Edwin L. Herr, Ed.D., is Distinguished Professor of Education, The Pennsylvania State University, University Park.