Nicola J. LoCascio, Marshall University, locascio@marshall.edu
Chris A. Barker, Marshall University
Robert Bickell Sr., Marshall University, bickel@marshall.edu

Abstract

The Biological Sciences Department at Marshall University was finding it difficult to offer the required courses for Biology majors as each curriculum stated. In addition, it was unclear if the course offerings truly represented the best training for the specialized majors. As a starting point for evaluating the curriculum we chose to measure current student satisfaction with introductory courses, preparation for upper-level courses and course offerings. Student satisfaction with their academic agenda has been linked to program strengths, and an assessment was undertaken to identify effective (and ineffective) areas of the curriculum as perceived by students. In particular, was the inclusion of defined core classes, currently set at three, giving students the necessary skills for their upper-level courses, and were there courses within the curriculum that contributed to success following graduation?

A survey questionnaire was developed and administered to students in eleven biological science courses. These courses spanned the biology curriculum and included the participation of freshman, sophomore, junior, senior, and graduate students A selected complement of controls embedded within the administered assessment tool, determined that being enrolled in all three core courses was the only variable that was statistically significant and a positive impact on the students’ perceptions of the quality of their scientific education. An outcome of this study revealed that completing, or currently being enrolled in, all three designated core courses resulted in students rating the quality of their scientific education significantly higher than students who have not taken all core courses. These findings were at the beginning of a restructuring of the Biological Sciences curriculum, with their associated required and recommended courses, and resulted in placing the core courses earlier in the academic programs.

Introduction

The curriculum at Marshall University Biological Sciences Department (BSC) was long overdue for evaluation regarding its effectiveness in preparing future Biologists. It was already noted that faculty attrition was negatively impacting the department’s ability to offer some required courses in a timely fashion. Recent faculty hires had changed the range of biological specializations present in our department resulting in some courses not being covered and new elective courses that were not specified for any of the six majors. An identification of current students’ perceptions was undertaken to help evaluate effective (and ineffective) areas of the curriculum. 

BSC originally offered six majors to undergraduate students: Biology; Botany; Environmental Biology; Microbiology; Physiology/Molecular Biology; and Zoology. All BSC undergraduates were required to successfully complete two prerequisite introductory courses and three core courses: Principles of Ecology, Principles of Cell Biology, and Principles of Genetics. Their remaining BSC credit hours were selected from upper-level classes that reflected their chosen specialization. The majority of students chose the Biology major, which, after the completion of the core classes, had no restrictions on the choices between remaining upper-level courses.  The traditional majors, considered strengths within the department, Botany and Zoology, lost most of their students to the general Biology major when the department was unable to offer the appropriate specialized upper-level courses for more than three years. Additionally, recent faculty hires had added new expertise in areas not represented by the current curricula and new course offerings that were not included as acceptable electives. It was past time to re-evaluate the BSC curriculum and determine its effectiveness to students.

It is arguable that students’ perceptions of collegiate experience are subjective in that they merely reflect the students’ feelings, and are influenced by their emotions. Administrators, as well as people outside of higher education, tend to consider student performance indicators such as standardized test scores, grade point averages, college retention, and graduation rates to be objective. Although these student performance indicators certainly measure various aspects of an institution’s success, they are unable to thoroughly provide significant information on the students’ perceptions, attitudes, and personal development of their educational experience (Cheng, 2001). Student outcomes such as satisfaction with academic courses, self-perception of knowledge and skills learned through course work, and their future ambitions are valid educational goals that should be used when performing an assessment (Stark et al., 1989; Corts et al., 2000). It is valid when investigating curriculum effectiveness to determine student opinions. It has been reported that involving students in this process can lead to better retention rates (Johnson, 1997), and measuring student satisfaction has identified areas for improvement in academic programs as a starting point for curriculum reform (Corts et al., 2000).

Two curriculum development workshops were convened for BSC faculty and staff. Discussions led to a decision to define individual majors with a restricted selection of classes that clearly defined each specialty as selected by faculty within each discipline. Courses defining the specialty majors were classified as either “required for major” or “allowable electives.” The required courses were to be offered annually, and the electives every two years, to ensure appropriate matriculation time for students. As this restructuring was approached, it became unclear whether all the traditional core classes were needed, or if only one or two would prepare students for their upper-level courses. Were the core classes really defining a BSC student, providing a solid, broad background, and paving the way for more intensive study? Should each Biological major have its own single core class that would track students into specialties during their sophomore year? Were there courses that delineate success in biological sciences, at our institution and in the work force? Certainly a core curriculum is designed to promote a more coherent set of learning goals (Ratcliff, 1992) but it was unclear if the BSC core was accomplishing their departmental and university mission.

This project was conceptualized according to what overall skills the university and biology department is committed to in educating its students. Also included were determinations for scientific and biological knowledge that a student is expected to attain. The assessment furthermore included the demographics of the students and opinions of what they believed were important for their general collegiate experience. Finally the assessment tool embedded the basic question: what factors contribute to determining differences among students with regard to their judgment of the quality of the scientific education they receive as biology majors?

Materials & Methods                          

Students responded to these questionnaire items using a scale. The scale consisted of five Likert items, commonly used in research of this kind (Pedhazur et al., 1991). The scale for both the general academic education skills and the biological science knowledge items was constructed using the following Likert items: (A) much stronger; (B) somewhat stronger; (C) a little stronger; (D) no change; and (E) diminished. The practicality scale used the following Likert items: (A) most important; (B) somewhat important; (C) a little important; (D) doesn’t matter; and (E) least important.

The questionnaire was administered to currently enrolled students of eleven undergraduate and graduate biological science courses for a total of 255 students. The questionnaires were then collected and the students’ responses were analyzed with the Statistical Package for the Social Sciences (SPSS) software package. SPSS permitted determination of significant differences between groups (e.g., t-tests, analysis of variance), and relationships among variables (e.g., correlation, multiple regression). The students’ responses to the demographic information were entered as they appeared on the questionnaire with the exceptions of sex, major, class status, ethnicity, and parents’ education level. Sex was entered as a “1” for male and a “0” for female. The student’ major was entered as a “1” for Biology, “2” for Chemistry, “3” for Math, “4” for non-science, and “5” for undecided. The student’s class status was entered as a “1” for freshman, “2” for sophomore, “3” for junior, “4” for senior, and “5” for graduate student. One high school student was enrolled in BSC 121; this was designated as a “0”. The student’s ethnicity was entered as a “0” for white or a “1” for other. The student’s parents’ education level was entered as a “1” for high school, “2” for Associate degree or some college, “3” for Bachelor degree, “4” for Master degree, and “5” for beyond a Master’s degree. Answers pertaining to the biological science prerequisite courses were entered as a “1” if the student had completed the course or was currently enrolled in the course. Courses not taken by a student were entered as a “0”. The final nineteen items containing the Likert-based responses were entered as a numeral corresponding to which item the student selected. A value of “5” was assigned to the response of “much stronger”; “4”.as assigned to “somewhat stronger”; “3” was assigned to  “a little stronger”; “2” was assigned to “no change”; and “1” was assigned to “diminished”. For the practicality of the students’ coursework section, a similar scale was assigned. A value of  “5” was assigned to the response of “most important”; “4” was assigned to “somewhat important”; “3” was assigned to “a little important”; “2” was assigned to “doesn’t matter”; and “1” was assigned to “least important”.

Included in the data collection were students enrolled in graduate BSC courses. The graduate courses had as pre-requisites the stated undergraduate core classes. Random coefficient regression and restricted maximum likelihood estimators were used in place of ordinary least squares estimators. This accommodates potentially confounding intra-class correlation and contributes to the accuracy of coefficient estimates and the validity of tests of significance. By choosing these analyses we corrected for any intra-class correlation from group commonality among students. This can occur when students in a group, by being together all semester, become more alike with regard to extraneous contaminating variables than students who are in different groups. Group commonality artificially deflates standard errors of regression coefficients, increases the probability that null hypotheses will be erroneously rejected, and inflates R2 values (Schroeder et al., 1986).

The assessment survey’s total sample size included the participation of 255 students. However, due to attrition or missing data on sixteen questionnaires, the sample size used for the statistical analysis of the assessment project totaled 239 students. After data had been collected, preliminary analyses were performed, and the research question became more specific and concrete: Does completion of all three core courses in the biology program increase students’ perception of the quality of scientific education received at Marshall University? (Core courses are BSC 320 Principles of Ecology, BSC 322 Principles of Cell Biology, and BSC 324 Principles of Genetics). Since the question is framed in a way that permits the answer as “improvement, yes or no”, the null hypothesis is that there has been no improvement in the students’ evaluation of the quality of scientific education at Marshall University and is the basis for statistical comparison. We are asking if the relationship between the number of our designated core classes the student has completed (CORE1, CORE2, or CORE3), and areas of scientific knowledge (SCIENCE), is readily attributable to random error, or if it actually reflects a non-zero relationship in the population of interest. Table 1 shows the coding for CORE1, 2 or 3.  

Table 1 - Composite Variable SCIENCE: Principal Component Loadings

Variance Explained = 58.2%
Cronbach’s Alpha = .82
N = 239

The composite variable SCIENCE incorporated survey questionnaire items that pertained to the students’ scientific education. These items correspond to questions 26 through 30 on the survey questionnaire (See Fig. 1). Collectively, these items reflect scientific knowledge gained through taking the biological science core courses. The areas of scientific knowledge included the students’ understanding of the scientific method, as well as cellular, genetic, molecular, ecological, and whole organism aspects of biology.

Factor analysis, combining the correlated variables into SCIENCE, was accomplished with principal component loading (see Table 2).  

Table 2- Composite Variable ACADEMIC:  Principal Component Loadings

Variance Explained = 60.7%
Cronbach’s Alpha (coefficient of reliability)= .86
N = 239

These items correspond to questions 19 through 25 on the survey questionnaire (Fig. 1). Collectively, these items reflect general academic skills attained by students during their collegiate experience. These areas included problem solving, writing, and critical thinking skills, the ability to synthesize information and plan complex projects, and understanding the role science has in society.

The other dependent variable, ACADEMIC, measured students’ perceptions of scientific skills drawn from their collegiate experience. Items used in constructing ACADEMIC are reported in Table 3, along with principal component loadings corresponding to each. The composite variable PRACTICL incorporated survey questionnaire items that pertained to the practical aspects of the students’ collegiate experience. Items used in constructing PRACTICL are reported in Table 4, along with principal component loadings corresponding to each.

Table 3 - Practical Education: Principal Component Loadings

Variance Explained = 65.4%
Cronbach’s Alpha = .73
N = 239

Student responses to questionnaire items 14-18 led to coding based on the number of BSC core courses in which the student was either currently enrolled or had completed. Biological science prerequisite courses were entered as a “1”. If no core courses had been taken then value was entered as a “0”.

These items correspond to questions 35 through 37 on the survey questionnaire. Collectively, these items reflect the students’ aspirations for attending college. These areas included such things as acquiring a broad educational background in the biological sciences, enhancing critical thinking skills, and understanding the relevance or practical value of biology coursework to real world issues.

The composite variables were defined according to three criteria: face validity, principal component configurations, and Alpha values. The face validity criterion is that the survey questionnaire items constituting the composite variables made sense. The items were intuitively appealing and consistent with everyday knowledge for participants in an academic setting. A principle component analysis was performed to further ensure the validity of the composite variables and all loadings were greater than the conventional cutoff value of .300. Finally, the Cronbach’s alpha criterion requires that all alpha values be at least .700, meaning that no more than thirty percent of the variability in a composite is due to random error. The independent variable of primary interest is CORE3 (Table 1), the variable that identifies students who have taken or are enrolled in all three core courses of the biology curriculum. In this instance, the alternative hypothesis is that the relationship is different from zero. Using null hypotheses and alternative hypotheses is conventional in applications of multiple regression analysis (Schroeder et al., 1986). If the unstandardized regression coefficient corresponding to CORE3 is statistically significant and positive, we will tentatively conclude that completion or enrollment in all three core courses, BSC 320, 322, and 324, in the biology program at Marshall University improves students’ assessment of the quality of scientific education offered at the University. Any other outcome will result in failure to reject the null hypothesis of no improvement.

Results

The demographics of the sample population were calculated from survey questions 1 through 4, 6 and 11 (see Figure 1) and closely matched the general student population of Marshall University as reported by the Marshall University Office of Institutional Research. The average value for sex was .41 indicating slightly more females than males in the sample, and ethnicity averaged .09, strongly skewed to all white respondents. The average class STATUS for the students was 2.95, or nearly junior standing (see Table 5). The average student’s parents’ education level, PARSED, was 2.78, nearly equivalent to a Bachelor’s degree, and the average MUGPA for the students was 3.40 and typically varied by .42 points (see Table 5).

N = 239

CORE1, CORE2 and CORE3 indicated if respondent had taken any of the three BSC core courses. STATUS measured respondent current class standing with high school level being coded 0 up to graduate level coded 5. The ACADEMIC variable was the sum of items for General Academic Education. SCIENCE was sum of items for Scientific Education and PRACTICL sum of items for Practical Education Expectations. These were combined from multiple questions on our questionnaire. SEX was coded 1 for male and 0 for female and ETHNIC coded 0 for white and 1 other. Parents’ education level, PARSED, was coded 1 for high school through 5 for beyond Masters level. Current grade point average, MUGPA, used a 4.0 scale rounded to nearest hundredth.

Table 5 lists the descriptive statistics with means and standard deviations. The composite variable SCIENCE could have a maximum value of 25, if a student reported that their scientific knowledge was much stronger as a result of taking specified courses. The minimum value for SCIENCE was 5, if a student reported that their scientific knowledge had diminished as a result of taking specified courses. The mean for SCIENCE was 19.67 and typically varied by 3.51 points. The composite variable ACADEMIC could have a maximum value of 30 if the student reported that their general academic education skills were much stronger as a result of taking specified courses, and a minimum value of 6, if the student reported that their general academic education skills had diminished as a result of taking specified courses. The mean for ACADEMIC was 23.16 and typically varied by 4.21 points. The composite variable PRACTICL could have a maximum value of 15, if the student reported that certain aspects of their collegiate experience were most important, and a minimum value of 3, if the student reported that certain aspects of their collegiate experience were least important. The mean for PRACTICL was 11.99 and typically varied by 2.24 points.

All independent variables except CORE1, CORE2, and CORE3, served as controls in multiple regression analyses (see Table 6). Having taken all three core classes (CORE3) might be strongly related to STATUS, simply meaning that students who have been at Marshall longer are more likely to have completed all three core courses. An association between CORE3 and the dependent variable SCIENCE (students’ assessment of scientific education) might be due to CORE3 being associated with STATUS. The composite variable SCIENCE incorporated survey questionnaire items that pertained to the students’ scientific education (Figure 1, items 26 - 30). Collectively, these items reflect scientific knowledge gained through taking the BSC core courses. To guard against making erroneous inferences due to the uncontrolled influence of STATUS, the SCIENCE variable was introduced as an additional independent variable. The unstandardized regression coefficient corresponding to CORE3 is statistically significant and positive (Table 6). In this instance, completing or being enrolled in all three core courses resulted in an increase of 1.12 points in the composite dependent variable, SCIENCE, relative to students who had taken none of the courses. The only other variable with a statistically significant coefficient is the ACADEMIC composite variable, which incorporated survey questionnaire items pertaining to the students’ general academic education (Figure 1, items 19 - 25).  Collectively, these items reflect general academic skills attained by students during their collegiate experience. These areas included problem solving, writing, and critical thinking skills, the ability to synthesize information and plan complex projects, and understanding the role science has in society (see Table 6).

Table 6 - Scientific Education as Dependent Variable. Random Coefficient Regression Results

R2L = 13.8%
ICC = 0.112     

Bold-Faced, Italicized Coefficients Statistically Significant, p<.05, One-Tailed Test

To further analyze the results, we combined CORE1, CORE2, and CORE3 into one variable, TOTCORE (Table 7). In this instance, instead of using three categorical independent variables, we used one variable, meaning that the variables can take on only the values of 0, 1, 2, or 3. As with Table 6, Table 7 shows that the more BSC core courses a student had completed, the higher the SCIENCE composite scale. Construing the independent variable in this way indicates that each course completed yields, on the average, a 0.48 point increase on the SCIENCE composite scale. The analysis does not enable us to distinguish among specific courses with regard to their independent effects on SCIENCE, but does show that taking core classes is cumulative, without regard to the order in which courses are taken.

R2L = 13.7%
ICC = 0.112    

Bold-Faced Coefficients Statistically Significant, p<.05, One-Tailed Test

Discussion

The results from the questionnaire administered to students enrolled in Marshall’s Department of Biological Sciences Department indicates that students are self-reporting significant gains in confidence in their ability to handle upper-level courses after they have experienced the required core courses. Students who have completed or are enrolled in the BSC core courses reported significant gains in academic skills, and an increased understanding of the scientific method, cellular, molecular, genetic, ecological, and whole organism biological aspects. These skills were seen as necessary by the faculty for success in the upper-level courses. By controlling for various factors such as students’ class status, gender, ethnicity, current grade point average, and parents’ education level, the only variables that made a significant difference in the students’ perception of the quality of scientific education were the defined core classes. Simply put, the three core courses are making a significant difference in the students’ satisfaction with their scientific education. The results also significantly demonstrate that the more core biology courses students have completed, the more the students favorably evaluate the quality of their science education.

Students self reported significant gains in problem solving skills, critical thinking skills, writing skills, public speaking skills, expanding their ability to execute complex projects, and understanding the role science plays in shaping our society. Students are benefiting from and appreciating their scientific education more through advanced biological coursework. On the average, students who are enrolled in or have completed all three core biology courses rated the quality of the scientific education they are receiving 0.48 points higher than students who had not completed the BSC core courses before enrolling in upper-level BSC courses.

The study did not find parents’ educational level to have a significant influence on the dependent variable SCIENCE. Parents’ educational background has previously been shown to be a predictor of a student’s academic success at the college level (Zhu, 2003). The level of parents’ education in the study ranged from 1.42 to 4.14; from just beyond high school up to the equivalent of a master’s degree. The average parents’ educational level reported was 2.78, or nearly a bachelor’s degree. The data reveal that students are attaining academic success in the core-centered curriculum regardless of their family’s educational background. This is particularly important at an institution such as Marshall, where many students will be the first in their families to complete a bachelor’s degree.  Educating students from families who have not traditionally completed secondary or post-secondary degrees has always been a challenge and priority at Marshall.  To have negated the potential influence of under-educated parents is a powerful statement for the BSC core curriculum.

Undergraduate core curricula, courses, are well established across American universities. Assessments of curriculum revisions are important for determining whether the mission of the institution is being met by a stated curriculum. It is equally important that revisions be attempted with faculty consensus, or they are less successful (Anderson et al., 1975). Periodic departmental curriculum reviews are also essential. The results of this assessment project reflect that students do indeed benefit from taking a defined set of core courses. In curriculum revisions, maintaining core classes can and does promote a more coherent set of learning goals (Ratcliff, 1992). For Marshall University’s Department of Biological Sciences, it was decided to maintain the present core classes and to add an additional core class (Principles of Microbiology) which then gave all our specialized majors an introductory level course. The Department will be monitoring this addition for its contribution to student satisfaction and preparation. The results of this study also resulted in enforcing student enrollment in core classes early in their collegiate years (at the sophomore or early junior level), rather than waiting until late junior and senior years. This study has determined that student satisfaction with their experience in the Department, in any of the biology majors, is significantly impacted by maintaining core classes for all students majoring in Biological Sciences.

Acknowledgement: The authors wish to thank Susan Weinstein for her critical reading of this manuscript.

References 

Anderson, S.B., S. Ball, R.T. Murphy, and Associates. (1975) Encyclopedia of Educational Evaluation, 1st ed. Jossey-Bass, Inc., Publishers, San Francisco, California. pp. 26-29, 311-318, 408-411, 458-462.

Cheng, D.X. (2001) Assessing Student Collegiate Experience: where do we begin? Assessment & Evaluation in Higher Education. 26: 525-538.

Corts, D. P., J. W. Lounsbury, R. A. Saudargas, and H. E. Tatum. 2000. Assessing Undergraduate Satisfaction with an Academic Department: A Method and Case Study. College Student Journal 34: 399–408.

Harvard-MIT Data Center’s Guide to SPSS. www.hmdc.harvard.edu/projects/SPSS_Tutorial/spsstut.shtml (March, 2004)

Johnson, A. 1997. “Assessment, Outcomes Measurement, and Attrition: Reflections, Definitions, and Delineations.” College and University 73: 14–17. 

Pedhazur, E., and L. Schmelkin. (1991) Chapter 2: Measurement and Scientific Inquiry in Measurement, Design and Analysis--An Integrated Approach. Lawrence Erlbaum Associates: Mahwah, New Jersey.

Ratcliff, J.L., (ed.) (1992) Assessment and Curriculum Reform: New Directions for Higher Education. No.80. Jossey-Bass Inc., Publishers, San Francisco, California, pp. 37-44.

Schroeder, L.D., D.L. Sjoquist,, and P.E. Stephan.. (1986) Understanding Regression Analysis An Introductory Guide. SAGE Publications, Newbury Park, California, pp. 11-34.

Schuh, J. H. and M.L. Upcraft. 2001. Assessment Practice in Student Affairs: An Applications Manual. San Francisco: Jossey-Bass.

Zhu, L. (2003) Who attains a bachelor’s degree in four years? Paper presented at the Northeast Association for Institutional Research annual conference, Newport, RI.