Correlating hardiness with graduation persistence.

Abstract

A five-campus longitudinal study of 1,432 first-year respondents found a positive correlation of students' "hardiness" with persistence to their expected '02 graduation, four years later. This capstone research's results reinforce two previous smaller studies that yielded similar outcomes. The report acquaints the reader with the hardiness construct, applies it to campus persistence efforts and outlines undergraduate retention policy ramifications inherent in such an approach. Specifically enrolling "lo-hardy" undergraduates in retention-intervention programs would provide a more efficient usage of expensive, labor-intensive retention intervention programs.

Introduction

Early identification of students who have a high potential to drop out of college has become a key concern of faculty and administrators involved with improving retention. Targeting at-risk undergraduates in retention-intervention programs provides a more financially efficient usage of expensive, labor-intensive, creative intervention programs. Many First-Year Seminars, for example, currently enroll all new arrivals to campus or only those students who voluntarily elect to participate. One challenge, therefore, for higher education administrators is to identify a mechanism that provides an early identification of these "at-risk" students. This report empirically tests the measurement of student hardiness as that mechanism.

The Research Response

This study is a rare use of hardiness as an independent variable in any longitudinal study. It replicates and reinforces two earlier published, more modest applications of hardiness as an independent variable linked to the dependent variable of undergraduate persistence to timely graduation within four years. The earlier studies involved 189 and 471 total arriving first year students respectively (Lifton & Flanagan, 1995; Lifton, et al., 2004). This current research replicated those earlier ones on five American campuses with 1,432 respondents.

In a preliminary attempt to find a user-friendly technique for identifying at-risk students, a pilot longitudinal study was undertaken at Ithaca College using a convenient, 30 item, multiple-choice hardiness questionnaire (Bartone, 1991). That study found a positive correlation between student hardiness and persistence to graduation (Lifton & Flanagan, 1995). The sample, however, was small and homogeneous in nature: 189 mostly white students all majoring in business disciplines. Further research was needed to determine if the approach is applicable to other settings.

A replication of the pilot study was undertaken at Widener University in Pennsylvania and Mississippi Valley State University in fall, 1997 with a larger, more diverse student population. The baseline results, published in AEQ, (Lifton, et al., 2000) launched a series of research inquiries that culminated with this current capstone study.

Four hundred seventy one (471) full-time, first-year, arriving students- somewhat equally distributed between the two campuses--answered the same hardiness measurement instrument used in the pilot study. Once again, the longitudinal analysis showed that persistence to their expected graduation in spring, 2001 correlated with hardiness (Lifton, et al., 2004). Lower-hardy students were disproportionately represented among the dropouts. That report also called for further research to confirm both the pilot effort and the two-campus findings. This current study, launched on five campuses with a virtually tripled sample size, 1,432 respondents, is in response to that call.

Describing Hardiness: A Literature Review

A substantial amount of psychological theory has focused on the human tendency to create and thrive during periods of change. Building on these schools of thought, Kobasa (1979) sought to describe a "hardy" person--one who welcomes and thrives during periods of stress.

Hardiness is a personality style that influences ways of thinking, feeling, and acting in the world that lead to personal growth rather than debilitation--particularly during times of stress. It is composed of three integrated components: commitment, challenge, and control. "Commitment" is a sense of value, meaningfulness, and loyalty towards one's purpose in life. "Challenge" fosters a willingness to leave behind the status quo to develop and grow with a new set of circumstances. "Control" is a belief in one's ability to influence events.

The reported research applications of hardiness are broad. Respondent groups have included nursing administration graduate students (Judkins, 2005), women survivors of sexual abuse (Feinauer, et al., 1996), children with attention deficit hyperactivity disorder (Brooks, 1994), farm and ranch families (Carson, et al., 1994), army disaster workers (Bartone, 1991), older adults (Magnani, 1990), executives (Maddi & Kobasa, 1984), bus drivers (Bartone, 1989), nurses (Harris, 1989), immigrants (Kuo & Tsai, 1986), student affairs administrators (Berwick, 1992), and adolescents (Hannah & Morrissey, 1987) among others.

Our research uniquely adds first-year students as a respondent group. We measured their hardiness just as they arrived on campus. Dependent variables in hardiness focused studies have included overall health and job stress (Sharpley, et al., 1999), war related stress (Bartone, 1999), drug use (Maddi, et al., 1996), basketball performance (Maddi & Hess, 1992), burnout (Pierce & Molloy, 1990), work performance (Westman, 1990) and general well being (Manning, et al., 1989; Lambert et al., 1989). Our research uniquely adds persistence to graduation as a longitudinal dependent variable. For some undergraduates, the transition to college life may prove far more stressful than exciting. Less able to manage the transition, these students may be more at risk of dropping out. Seidman (1996), for example, urges college and university administrators to use the data at their institution's disposal to develop student profiles of those who, historically, do not persist to graduation.

Our approach expands Seidman's campus-by-campus focus to a more comprehensive attempt applied to undergraduates in different settings. It emphasizes measuring student hardiness as they begin the higher education enterprise. Targeting "lo-hardy" students, regardless of their background, for enrollment in retention intervention programs might permit more efficient usage of these expensive, labor-intensive efforts that are best suited for those who truly need them.

Measuring Hardiness

As interest in the construct develops, hardiness methodologists will need to reach some consensus on how to assess it. Hoping to create an "... improved measure that is grounded conceptually in the original work on the hardiness construct but ... corrects the psychometric limitations of the earlier measure(s)," Bartone (1991, p. 2) developed, tested, and reported the use of a 30-item instrument in 13 samples. Recently, an even briefer 18-item survey shows great promise in recent test environments (Maddi & Khoshaba, 2001).

Bartone's instrument, slightly reworded for an undergraduate respondent base, was distributed to 1,432 full time, first year students early in the fall '98 semester: 198 (13.8% of the total) at Mississippi Valley State University (MVSU); 92 (6.4% of the total) at Texas A&M University at Kingsville; 512 at Mississippi State University (35.8% of the total); 443 at Elon College in North Carolina (30.9% of the total); and 187 at Pacific Lutheran University in Washington state (13.1% of the total). Demographic data about each respondent were added to the data sets.

The Baseline Findings

The quite diverse racial/ethnic distribution of the sample, displayed in Table I, includes those who purposefully chose not to disclose it in comparison to data that is outright "missing." See issue website http://rapidintellect.com/AEQweb/fal2006.htm Hardiness levels in the current study were randomly distributed across these racial/ethnic categories. This repeated a finding from the prior two-campus study. Gender, however, proved to be a different matter. The current sample's mean hardiness score was 56.49 across the five campuses with a standard deviation of 7.36. Female students in the sample averaged almost two points higher than males in both this study and its predecessor (See Table II). See http://rapidintellect.com/AEQweb/fal2006.htm This finding, now noted in two separate samples, contradicts the conventional wisdom that hardiness is unrelated to gender. It may be that within the niche of graduating American high school seniors, lo-hardy girls are less eager to take on the demands of college.

Finally, it is noteworthy to turn the baseline analysis to a discussion of two popular measures that are used to evaluate American student college admission portfolios: High school rank in class and the SAT and ACT national college admissions tests. Hardiness did not correlate with either national test but, as Table III demonstrates, did correlate with rank in class at the .006 level of statistical significance. See issue website http://rapidintellect.com/AEQweb/fal2006.htm Thus, while hardiness does not help measure one's (academic) potential (ACT/SAT), it may facilitate one's ability to rise to it (H.S. rank in class; becoming an alum).

The Longitudinal Results--Who Is At Risk?

Before focusing on the fundamental thrust of this research--confirming a test for hardiness for the third time as a mechanism to identify at-risk students--it is appropriate to examine some obvious alternatives to hardiness as the indicator: national admissions tests (SAT and ACT); and high school rank in class. Those in the sample who dropped out tended to have statistically significant lower SAT or ACT scores than those who persisted to graduation. Test scores of students continuing to register beyond their projected graduation year averaged national test scores in between those who graduated on time and those who dropped out. It will become less appealing, however, to target low scoring SAT/ACT students as at-risk as campus admissions officers across the United States continue to debate the efficacy of using these national tests as an admissions criterion.

Persistence rates to graduation year in this current sample also correlated at a .001 level with high school rank in class. Half our sample came from the top quartile of their high school graduating class. Four years after enrolling on campus, 43% graduated, 29% continued as registrants while 28% dropped out. The statistics for the 31% of the sample in the second high school quartile are 35% graduated, 28% continuing and 37% as dropouts.

The results are more disheartening in the lower half's two quartiles, 14% and 5% of the overall sample respectively. The third quartile's outcome status is only 23% graduated, with 31% continuing and 46% as dropouts. The bottom quartile's numbers are 25%, 33% and 42% respectively. Lacking a better tool, retention administrators who are looking to target at-risk students for their programmatic interventions might therefore consider high school class rank. One problem using rank in class as an indicator of being an at-risk student is in the absolute numbers involved. Since 50% of the respondents were not in their high school's first quartile, this may generate too large a number to target given a possible scarcity of retention resources to allocate on campus. With only five percent of the total sample in the fourth quartile alone, this may yield too few students.

Now, hardiness. To review, the sample's mean hardiness score across the 1,432 usable student responses who arrived on campus in F'98 was 56.49. We successfully tracked 1,415 of these respondents to their graduation year. As shown in Table IV, those who graduated on time were the only group who averaged above the sample's mean hardiness score. Those who dropped out had the lowest hardiness score with continuing students registering a hardiness score in between the other two groupings. See issue website http://rapidintellect.com/AEQweb/fal2006.htm

Summary and Conclusions

To review, our most noteworthy finding was the statistically significant correlation of hardiness with persistence to the undergraduates' expected graduation four years out. These results echo outcomes in the study's two previous iterations reported earlier (Lifton & Flanagan, 1995; Lifton, et al., 2004). Moreover, this research also found that while hardiness correlated with high school rank in class, it did not correlate with national entrance examination scores, the SAT and ACT. It seems that the hardiness construct can be seen as an indicator of students' ability to meet their potential, i.e. high school grades, without indicating what that potential is, national examination scores.

Along the way, other empirical results of this four-year longitudinal study may interest hardiness scholars. Once again, hardiness was not found to correlate with ethnicity/race variables suggesting all levels of hardiness are found equally among all such backgrounds. The reported correlation here of hardiness with sexual category--females testing higher-hardy than males--does confound the general understanding that the construct is gender free, at least among the college-applicant subset.

With the publication of this article, the correlation of undergraduate hardiness with undergraduate retention to expected graduation year has been demonstrated again, this time among 1,432 diverse, arriving first-year students distributed among five campuses. This capstone research may give retention administrators confidence to adopt hardiness testing as a user-friendly tool that identifies at-risk students upon their arrival. It makes sense, financially, to target lo-hardy students for campus labor-intensive, expensive retention intervention programs while exempting others from the drill.

One could suggest that a hardiness retention model might be used instead as an admissions criterion to weed out possible future dropouts altogether from the entering class. This seems unlikely on at least three counts: (a) Ethically, it runs counter to the undergraduate instruction mission of American higher education institutions; (b) pragmatically, it could become a public relations nightmare for any college if their use of such an admission screen leaked out; and (c) cynically, the need for "making the numbers" on a sufficient quantity of new enrollees to contribute their tuition dollars to annual campus operating budget revenues makes it financially far more prudent to admit lo-hardy students and then find ways to help them achieve their graduation goals (as they continue to persist as tuition payers).

This capstone report marks the third successive time that a longitudinal study found a positive correlation of undergraduate hardiness with persistence to expected graduation four years later. Campus administrators might now consider measuring undergraduates' hardiness upon their arrival as a proven procedure to identify which new students need the nurturing attention of the expensive retention intervention programs already shown to help undergraduates persist to graduation. Focusing immediate retention efforts--perhaps as a pilot project--exclusively on the lowest hardy students may prove to be a helpful and cost effective interim measure until more transformational campus reform can be implemented.

Note

An earlier version of this paper was presented at the 2005 Annual Meeting of the American Society of Business and Behavioral Sciences (ASBBS) and appears in its proceedings.

References

Bartone, Paul (1989). Predictors of stress-related illness in city bus drivers. Journal of Occupational Medicine, 31, 657-663.

Bartone, Paul (1991, August). Development and validation of short hardiness measure. (Poster presentation at the 3rd Annual Convention of the American Psychological Society, Washington, DC. Author available at National Defense University.)

Bartone, Paul (1999). Hardiness protects against war related stress in army reserve forces. Consulting Psychology Journal, 51(2), 72 83.

Berwick, Kathleen. (1992). Stress among student affairs administrators: The relationship of personal characteristics and organizational variables to work related stress. Journal of College Student Development, 33(1), 11 19.

Brooks, Robert B. (1994). Children at risk: Fostering resilience and hope. American Journal of Orthopsychiatry, 64(4), 545 553.

Carson, David K., Araquistain, Mary, Ide, Betty, Quoss, Bernita, & Weigel, Randy (1994). Stress, strain, and hardiness as predictors of adaptation in farm and ranch families. Journal of Child and Family Studies, 3(2), 157 174.

Feinauer, Leslie L., Mitchell, Jill, Harper, James M., & Dane, Skip (1996). The impact of hardiness and severity of childhood sexual abuse on adult adjustment. American Journal of Family Therapy, Fall, 24, 206 214.

Hannah, T. Edward, & Morrissey, Craig (1987). Correlates of psychological hardiness in Canadian adolescents, Journal of Social Psychology, 127, 339-344.

Harris, Ruth B. (1989). Reviewing nursing stress according to a proposed coping-adaptation framework. Advances in Nursing Science, 11, 12-28.

Judkins, S. K., Arris, L., & Keener, E. (2005). Program evaluation in graduate nursing education: Hardiness as a predictor of success among nursing administration students. Journal of Professional Nursing. 21 (5), 314-321.

Kobasa, Suzanne C. (1979). Stressful life events, personality and health: An inquiry into hardiness. Journal of Personality and Social Psychology, 37, 1-11.

Kuo, Wen H., & Tsai, Yung-mei (1986). Social networking, hardiness and immigrant's mental health. Journal of Health and Social Behavior, 27, 133-149.

Lambert, Vickie A., Lambert, Clinton E., Klipple, Gary L., & Mewshaw, Elizabeth A. (1989). Social support, hardiness, and psychological well-being in women with arthritis. Image: Journal of Nursing Scholarship, 21, 128-132.

Lifton, Donald E. & Flanagan, Leo (1995). Student "hardiness" as a predictor of undergraduate persistence. Proceedings of the Fourth Canadian-American Conference on the First-Year Experience, Columbia, South Carolina, The Freshman Year Experience, 66-67.

Lifton, Donald E., Seay, Sandra & Bushko, Andrew (2000). Can student "hardiness" serve as an indicator of likely persistence to graduation? Baseline results from a longitudinal study. Academic Exchange Quarterly, 4, 73-81.

Lifton, Donald E., Seay, Sandra & Bushko, Andrew (2004). Measuring undergraduate hardiness as an indicator of persistence to graduation within four years in Duranczyk, I. M., Higbee, J. L., Lundell, D. B. (Eds) Best Practices for Access and Retention in Higher Education, Minneapolis, MN: Center for Research on Developmental Education and urban Literacy, General College, University of Minnesota, 103-113.

Maddi, Salvatore R., & Hess, Michael J. (1992). Personality hardiness and success in basketball. International Journal of Sport Psychology, 23(4), 360 368.

Maddi, Salvatore R., & Khoshaba, Deborah M. (2001). PVSIII-R: Test development and internet instruction manual. Newport Beach: The Hardiness Institute.

Maddi, Salvatore R., & Kobasa, Suzanne C. (1984). The hardy executive: Health under stress. Homewood, IL: Dow Jones-Irwin.

Maddi, Salvatore R., Wadhwa, Pathik, & Haier, Richard J. (1996). Relationship of hardiness to alcohol and drug use in adolescents. American Journal of Drug and Alcohol Abuse, 22(2), 247 257.

Magnani, Lorraine E. (1990). Hardiness, self-perceived health, and activity among independently functioning older adults. Scholarly Inquiry for Nursing Practice: An International Journal, 4, 171-184.

Manning, Michael R., Williams, Rudy F., & Wolfe, Donald M. (1989). Hardiness and the relationship between stressors and outcomes, Work & Stress, 2,205-216.

Pierce, C. Mark B., & Molloy, Geoffrey N. (1990). Psychological and biographical differences between secondary school teachers experiencing high and low levels of burnout. British Journal of Educational Psychology, 60, 37-51.

Seidman, Alan (1996). Retention revisited: RET = E Id + (E + I + C)Iv. College and University, 71 (4), 18-20.

Sharpley, Christopher F., Dua, Jagdish K., Reynolds, Roisin & Acosta, Alicia (1999). The direct and relative efficacy of cognitive hardiness, a behavior pattern, coping behavior and social support as predictors of stress and ill health. Scandinavian Journal of Behavior Therapy, 1, 15 29.

Westman, Mina (1990). The relationship between stress and performance: The moderating effect of hardiness. Human Performance, 3, 141-155.

Donald Lifton, Ithaca College, NY

Sandra Seay, East Carolina University, NC

Nancy McCarly, Mississippi State University, MS

Rebecca Olive-Taylor, Elon University, NC

Richard Seeger, Pacific Lutheran University, WA

Dalton Bigbee, Texas A&M University at Kingsville, TX

Lifton is an Associate Professor of Management, Seay is an Assistant Professor of Education, McCarly directs her university's honors program, Olive-Taylor coordinates her university's tutorial services, Seeger is a Senior Advisor in his university's Student Academic Success program, Bigbee is his university's Associate V.P. of Academic Affairs