• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

The relationship among academic performance, age, gender, and ethnicity in distance learning courses delivered by two-year colleges

Dissertation
Author: Bruce Jost
Abstract:
This study investigated the effects the demographic variables age, gender, and ethnicity and their interactions had on academic performance in distance learning courses delivered by public two-year colleges in Kentucky. The study controlled for previous academic performance measured by cumulative grade point average (GPA). The study used a random sample ( N = 320) of all students who had enrolled in at least one distance learning course delivered by the institutions of the Kentucky Community and Technical College System in the spring 2008 semester. A linear hierarchical multiple regression acting as ANCOVA served as the main analysis, with the order entry: cumulative GPA, independent variables (age, gender, ethnicity), interaction vectors, product vectors. Final course grade served as the dependent variable. The results of the analysis indicated that only cumulative GPA was a significant predictor, explaining approximately 40% of the variance of the final grade. Although differences in final grades were present among the variables age and ethnicity, these differences disappeared when controlling for cumulative GPA. Significance of the results and recommendations for future research are discussed.

Research Setting The site of the study was Kentucky, which has experienced a large and rapid growth in the last decade of distance learning courses offered by its community and technical colleges. Each of the 16 colleges of the KCTCS offer distance learning courses in a variety of disciplines. The courses are implemented in the Blackboard Academic Suite, a comprehensive software package designed to deliver academic courses over the World Wide Web. Blackboard allows delivery of content and creation and grading of assessments such as tests. Each course also has a communications center that allows students and instructors to communicate via e-mail, discussion boards, or chat sessions. Instructors have the ability to provide documents, audio, and video content to their courses. Students can track their progress with an electronic gradebook. Distance learning courses follow one of two time formats. Approximately 19 courses are offered as Learn Anytime courses. These courses are open enrollment and students may enroll in them at any time of the year: they are not bound by the academic calendar. Most of the distance learning courses, however, follow the academic semester calendar set by the KCTCS. This study only included the courses that followed the semester calendar. The tuition for distance learning courses is the same as the in-state tuition for on-campus courses. Although the implementation of distance learning courses varies with each instructor, the vast majority of the courses are mostly asynchronous. Projects or homework may have deadlines and the instructors may require taking a test within a certain time window, but otherwise students can work on their course anytime and 9

anywhere. Some of the instructors may require proctored exams where students must visit their local college. Courses with labs courses may require students to visit a local college a few times in the semester to perform labs that would otherwise be difficult or impractical to perform at home. Limitations This study was subject to certain limitations. Results of the study could have limited generalizability because of the sample used. The possibility exists that the results of the study are applicable to only distance learning courses delivered by public two-year colleges in Kentucky. The study considered only one type of distance learning (online delivery), which also limited the generalizability of the results. The sample was from only one semester, which may further limit generalizability of the results. The number of ethnic groups represented in the sample was small and the data had to be aggregated into three groups: White, Black, and Other/Not Reported. The study employed final course grade as the measure for academic performance. Some researchers have suggested that course grades may not be the best measure of academic performance (Astin, 1993), implying that the results of the study might not have been measuring what they intended to measure. That is, a problem with the validity of the dependent variable might have been present. A possibility existed that many of the failing grades in the sample were a result of participants not completing the course and not officially withdrawing from the course. In other words, these participants had not truly failed the course, but instead did not follow administrative procedures. This practice also threatens the validity of the dependent variable and may skew its distribution 10

because if these students had followed administrative procedures and withdrawn from their courses, the number of failing grades probably would have been lower. The study made use of existing student data in the KCTCS database, which generally reduces the threats to validity and reliability of the data. Some of the data, however, was self-reported from the participants from applications for admission. Staff at one of the KCTCS colleges then entered the information from the applications into the system database of student records. The possibility existed that the participants did not report accurate age, gender, or ethnicity information or that the staff made errors when inputting the information. Similar potential sources of error were present with the reporting of grades, which threatened the accuracy of the dependent variable and control variable (i.e., cumulative GPA). Summary of Chapter This chapter presented the rationale for the current study. Enrollment in distance learning courses and programs has been increasing by double-digit percentages for most of the current decade. Institutions are incorporating distance learning into their strategic plans and plans for growth. Many researchers and practitioners are looking to distance learning as a democratizing force in higher education (Allen & Seaman, 2007; Van Dusen, 2000). A need exists to ensure that certain groups of students do not underperform in distance learning coursework because of additional barriers erected due to the nature of distance learning. The first step towards this assurance is to determine whether certain groups of students actually do underperform. This study attempted to take this initial step. The purpose of the study was to determine the relationship between academic 11

performance in a distance learning course delivered by a two-year college in Kentucky and the demographic variables of age, gender, and ethnicity. The following chapter reviews previous research on the research problem of the current study. 12

CHAPTER II REVIEW OF LITERATURE This chapter contains a review of studies related to academic performance in postsecondary education. The purpose of the literature review was to: (a) review the effects of the main variables of interest (age, gender, ethnicity) on academic performance; (b) identify other factors that may influence academic performance; and (c) review the methodologies of previous studies. The results of these three objectives will ground the study design and help explain the findings. The purpose of the study was to investigate the relationship between academic performance in distance learning courses delivered by two-year colleges and the demographic variables of age, gender, and ethnicity. Four research questions guided this study: what are the relationships between the three main variables and their interactions and academic performance? Ideally, this literature review would restrict its focus to only those studies related to distance learning. Because of the paucity of such studies, however, this chapter must include reviews of studies dealing with academic performance in on-campus courses. This chapter contains four main sections. The first section reviews studies related to age and includes two subsections: performance in on-campus coursework and 13

performance in distance learning coursework. The second main section reviews the relationship of gender and ethnicity to academic performance. The third main section briefly reviews studies that identified other factors significantly related to or predictive of academic performance. The fourth main section reviews models of academic performance to provide a conceptual framework for the study. A summary concludes the chapter. Age and Academic Performance The nontraditional-aged student became a subject of interest in higher education after World War II when large numbers of returning veterans started using their benefits guaranteed in the Servicemen's Readjustment Act of 1944 (i.e., the G.I. Bill). Institutes of higher education (IHEs) started experiencing a wave of students who did not fit the traditional college student profile. The veterans were older, had interrupted their education after high school, and many had disabilities. At first, faculty at IHEs viewed the older student as having a deficiency of performance in undergraduate studies (Kasworm, 1990). Much of the early research addressed this implied deficiency or the older students' supposed needs (Richardson & King, 1998). The current subsection reviews studies from researchers who tried to determine if older students truly did underperform traditional-aged students in on-campus coursework. Studies ofOn-Campus Learning and Age Bellico (1972) examined the student characteristics that could predict performance in a series of economics courses. The participants (N= 92) were students who completed a sequence of advanced economic courses in their junior and senior years 14

at a university in the Northeast. The article did not mention the sampling procedures. The dependent variable was the participant GPA in the advanced economics courses. The researcher used 18 independent variables including age, gender, GPA of freshman and sophomore year courses, attendance at a community college (nominal scale: yes or no), high school GPA, and Scholastic Aptitude Test (SAT) scores. When computing the correlation coefficients, the researcher found that 12 of the independent variables were significantly (p < .05) correlated with the dependent variable. Freshman and sophomore GPA had the largest correlation, r = .67. A stepwise multiple regression found that two independent variables accounted for most of the variation in the model. Freshman and sophomore GPA and attendance at a community college explained 56.7% of the variance in GPA of the advanced economics courses. Freshman and sophomore GPA was positively related and attendance at a community college was negatively related so that those participants who attended a community college were more likely to do worse in the economics courses. Age was not a significant predictor of performance. Houltram (1996) examined differences in academic performance between students in different age groups in a pre-registration nursing program in the United Kingdom (U.K.). The participants (N = 225) were students enrolled in the Common Foundations Programme (CFP), which is the first phase of the nursing education curriculum in the U.K. The researcher did not perform any sampling. The main independent variable was age, categorized as either traditional (17-21 years, n = 132) or nontraditional (22 years and older, n = 93). Entry mode was an independent variable and nominal with two levels and described whether the participant entered the program as academically qualified (i.e., passed at least five General Certificate of Secondary 15

Education, or GCSE, exams) or with alternative qualifications (i.e., passed the DC Test). The dependent variable was academic performance, which consisted of two parts: the overall mean score for the CFP for each student (interval scale); and success, which became a nominal variable with two levels ("successful" = one or no low grades, "less successful" = two or more low grades and/or repeats and/or dropped program). The researcher used an independent-samples t test with the whole sample with the mean CFP score as dependent variable and age group as the independent variable. The test revealed a significant difference, r(185) = -4.49, p < .01, between the two groups with the older group having higher scores. The researcher then repeated the analysis separately on those participants who entered with academic qualifications and those with alternative qualifications. The t tests revealed that regardless of entry mode a significant difference (p < .01) between the two age groups existed with the older participants obtaining higher scores. Darkenwald and Novak (1997) examined the relationship between the proportion of nontraditional-aged students in a class and the academic performance of the students in that class. The participants (N= 2,794) were students enrolled in evening or weekend classes at either a suburban community college or a large research university. The researchers performed parallel studies at the two sites and reported the results separately. The community college sample comprised 44 randomly sampled classes and used all enrolled students in those classes in = 619). The university sample (n = 2,175) was not random but instead consisted of the students in the 28 classes that contained over 50% of nontraditional-aged student enrollment as well as the students in 28 classes that had 24%- 49% nontraditional enrollment and the students in 28 classes that had less than 25% 16

nontraditional enrollment. The researchers selected the 56 classes in the latter two categories to match the academic discipline of the 28 high adult-enrollment classes. The dependent variable was academic performance measured with the final course grade and converted to a GPA. The independent variable was class age composition, which was nominally scaled with three levels. This variable indicated the proportion of nontraditional-aged students in a class. A class was predominantly adult if at least 60% of the participants in that class were over 24 years old. The class was predominantly traditional-aged if at least 60% of the participants in that class were under 24 years old. Any class that had between 41% and 59% nontraditional-aged participants was split age. For the university sample, the researchers changed the percentages of nontraditional-aged participants to 50% or more for predominantly adult, 25%-50% for split age, and less than 24% for predominantly traditional-aged. Age served as a control variable and was nominal scaled with three levels (pre-adult, 23 years or younger; young adult, age 24-29; and mature adult, 30 or older). The researchers performed a 3 (class age composition) x 3 (age group) ANOVA for each of the two samples. For the community college sample, the analysis revealed that only the age composition main effect was significant,/* < .01. A Scheffe post hoc analysis revealed that the participants in the predominantly adult classes scored significantly higher than those in the other two groups. In addition, the participants in the split age classes scored significantly higher than those in the predominantly traditional- aged classes. The effect size was 25%. 17

The initial ANOVA for the university sample did not produce a significant result. After eliminating the participants enrolled in mathematics classes, the ANOVA revealed a significant age composition main effect,/? < .01. The Scheffe post hoc analysis indicated that the participants in the predominantly adult classes had scored significantly higher than those in the other two groups. In addition, the participants in the split age classes scored significantly higher than those in the predominantly traditional-aged classes. Peiperl and Trevelyan (1997) examined which student characteristics (e.g., standardized test scores, age, gender, etc.) predicted student performance in an international Masters of Business Administration (MBA) program at a university in the U. K. The participants (N= 362) were first-year students in the MBA program. The researchers performed no sampling. The dependent variable was academic performance as measured using final course grades, grades on individual assignments, and grades on group assignments. The independent variables were age of the student (interval scaled), gender, marital status (levels not reported), years of full-time work experience (interval scaled), language proficiency (three levels: native English speaker, non-native English speaker/educated in English, and non-native English speaker/not educated in English), and verbal and quantitative scores on the Graduate Management Admission Test (GMAT), which was a requirement for entry into the program. The researchers performed three separate stepwise multiple regression analyses using each of the three performance measurements as dependent variables and the independent variables as predictors. The tests revealed that GMAT verbal score was the strongest predictor of final and individual assignment grades in the positive direction. 18

Full document contains 118 pages
Abstract: This study investigated the effects the demographic variables age, gender, and ethnicity and their interactions had on academic performance in distance learning courses delivered by public two-year colleges in Kentucky. The study controlled for previous academic performance measured by cumulative grade point average (GPA). The study used a random sample ( N = 320) of all students who had enrolled in at least one distance learning course delivered by the institutions of the Kentucky Community and Technical College System in the spring 2008 semester. A linear hierarchical multiple regression acting as ANCOVA served as the main analysis, with the order entry: cumulative GPA, independent variables (age, gender, ethnicity), interaction vectors, product vectors. Final course grade served as the dependent variable. The results of the analysis indicated that only cumulative GPA was a significant predictor, explaining approximately 40% of the variance of the final grade. Although differences in final grades were present among the variables age and ethnicity, these differences disappeared when controlling for cumulative GPA. Significance of the results and recommendations for future research are discussed.