08832323.2013.836470
Journal of Education for Business
ISSN: 0883-2323 (Print) 1940-3356 (Online) Journal homepage: http://www.tandfonline.com/loi/vjeb20
Business Statistics: A Comparison of Student
Performance in Three Learning Modes
Gerald R. Simmons
To cite this article: Gerald R. Simmons (2014) Business Statistics: A Comparison of Student
Performance in Three Learning Modes, Journal of Education for Business, 89:4, 186-195, DOI:
10.1080/08832323.2013.836470
To link to this article: http://dx.doi.org/10.1080/08832323.2013.836470
Published online: 02 May 2014.
Submit your article to this journal
Article views: 140
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=vjeb20
Download by: [Universitas Maritim Raja Ali Haji]
Date: 11 January 2016, At: 20:39
JOURNAL OF EDUCATION FOR BUSINESS, 89: 186–195, 2014
C Taylor & Francis Group, LLC
Copyright
ISSN: 0883-2323 print / 1940-3356 online
DOI: 10.1080/08832323.2013.836470
Business Statistics: A Comparison of Student
Performance in Three Learning Modes
Gerald R. Simmons
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Texas A&M University-Central Texas, Killeen, Texas, USA
The purpose of this study was to compare the performance of three teaching modes and age
groups of business statistics sections in terms of course exam scores. The research questions
were formulated to determine the performance of the students within each teaching mode, to
compare each mode in terms of exam scores, and to compare exam scores by age group. The
research hypotheses predicted there would be a difference between the three teaching modes
and age groups. The results found significant differences in both the online teaching mode for
exam 3 and age groups for the mean of the three exams.
Keywords: blended, business statistics, face-to-face, Hawkes Learning Systems, online, teaching modes
As we fast approach the closing of the first quarter of this century, methods of receiving higher level education continues
to change from the traditional mode of university learning
in face-to-face (F2F) lectures and labs. Online learning is
quickly becoming the new traditional method of learning.
Radford (2011) found that between 2000 and 2008 the percentages of undergraduates enrolled in at least one distance
education class expanded from 8% to 20%. Additionally,
the percentage of undergraduates enrolling in distance education programs increased from two to four percent during
the same period (Radford, 2011). Of all students, Radford
also found 24% of the business students enrolled in distance
education courses and 6% enrolled in distance education
programs. In terms of enrollment, the 2009 figures show
University of Phoenix, Online Campus, has the highest enrollment of all U.S. postsecondary institutions, with 380,232
students (National Center for Education Statistics, 2011).
Kaplan University is second, with 71,011 students, and Arizona State University is third with 68,064 students (National
Center for Education Statistics, 2011). The other two institutions that made the top five were Miami Dade College
at 59,120 students and Ohio State University, Main Campus, at 55,014 students. The University of Minnesota, Twin
Correspondence should be addressed to Gerald R. Simmons, Texas
A&M University-Central Texas, Department of Management and Marketing, 1001 Leadership Place, Founders’ Hall, Killeen, TX 76549, USA.
E-mail: [email protected]
Cities, was rated ninth; the University of Texas at Austin
was rated 10th; Texas A&M University was 12th; and the
University of Washington, Seattle Campus was rated 18th
(National Center for Education Statistics, 2011). What is interesting to note is the fact the top two institutions are fully
online universities. Additionally, most of the listed traditional
universities either have fully online or fully distance degree
programs.
This regionally accredited university provides it students
with upper level (baccalaureate) and graduate programs and
provides both online, blended, and traditional courses and
degree programs. The objective of this study was to expand
the current knowledge concerning the perceptions of online
learning versus the perceptions of traditional learning; therefore, this study presents a comparison of online, blended, and
traditional F2F business statistics sections at a single university. The purpose of this quantitative study was to compare the
performance of F2F, blended, and online business statistics
sections in terms of course exam scores, controlling for age,
gender, previous experience in statistics, and course length,
of the sections that met between the fall of 2010 and the summer of 2012 at a regionally accredited university. This study
was differentiated from other online, F2F, and blended comparison studies in that the data collected were from sections
that were taught using the same textbooks, as in the studies
by Dotterweich and Rochelle (2012) and McLaren (2004),
but also administered the same assignments and exams in the
same manner; thus reducing possible instructor bias. These
details will be explained in the following discussion of the
Hawkes Learning System (HLS)
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
BUSINESS STATISTICS: A COMPARISON
This study differed from Dotterweich and Rochelle whose
online students completed online assignments while the F2F
students completed the traditional assignments from the textbook. This study differed from McLaren (2004), whose F2F
and online assignments were also presented in a different
form and the exams were administered differently. F2F students could take their exams with a sheet of notes, whereas
the online students had access to all notes and textbooks, with
the request they do not receive outside assistance (McLaren,
2004). Additionally, this study sought to determine the effects of the course length. The university taught the courses
in eight and sixteen week sessions. The following sections
provide a brief background of online learning, the online
and F2F pedagogies of the regionally accredited university’s
business statistics courses, methodology, analysis, and finally
the results of the study.
BACKGROUND
Online Learning
The roots of distance education and online learning stemmed
from correspondence courses. These courses were taught
over a geographic distance where instructors and students
corresponded through the postal system. With the introduction of the Internet and the world wide web, in 1989, online
and distance education began its expansive growth. It was
at this time that institutions, such as University of Phoenix,
established their online programs (University of Phoenix,
2012). The advantage of the Internet and world wide web
was the quick transmission and correspondence between the
instructors and the students. Online learning grew from email
transmissions, to newsgroups, to the online learning management systems (e.g., Blackboard, Moodle, Desire2Learn).
Online Versus Blended Versus Traditional
Learning
Online learning consists of students receiving instruction,
conducting research, completing assignments, quizzes, and
exams, and interacting with the instructor in a format that uses
the world wide web as the basis for a learning management
system. This interaction can be either in a synchronous mode,
or in an asynchronous mode (Mills & Raju, 2011; Summers,
Waigandt, & Whittaker, 2005; Tallent-Runnels et al., 2006).
Blended learning courses consists of traditional F2F lecture
periods, coupled with online components such as group or
team study, assignments, quizzes, and exams (Mills & Raju,
2011). Traditional learning consists of the F2F interactions
according to scheduled periods, where students receive information (or lectures) in a classroom.
The idea that traditional learners perform better than online learners has pervaded the thoughts of professionals and
academicians since the advent of distance learning. This idea
became more prevalent with the growing numbers of ac-
187
credited online universities and their growing populations of
students and the online programs within traditional universities, spurring many studies comparing the effects of online
learning and traditional learning (DeNeui & Dodge, 2006;
Harrington, 1999; Mills & Raju, 2011; Summers et al., 2005;
York, 2008). Studies of comparison sought to determine if
there were true differences in student learning between online
and F2F (Bernard et al., 2004; Harrington, 1999; Summers
et al., 2005), blended and online (Arbaugh, Desai, Rau, &
Sridhar, 2010), and all three (Ashby, Sadera, & McNary,
2011; York, 2008). The results of these studies included a
poorer performance in the traditional classroom (Ashby et al.,
2011) as well as no significant difference in the three modes
of instruction (Harrington, 1999; Summers et al., 2005; York,
2008). All researchers recommended further study, especially
with the expected increase of online learning.
Business Statistics and Hawkes Learning
System
Technology has always taken a role in the education of students, beginning with the abacus and continuing on to the
laptop and notebook computers of today. Without the use of
technology, courses such as statistics became the dread of
many students whose intent was to learn only the basics, not
actually major in the subject.
Technologies used in many introductory statistics courses,
regardless of teaching mode primarily consisted of some
form of computational software. These software packages
included the Texas Instruments 84 (Dallas, TX) family of
pocket calculators, Microsoft Excel (Seattle, WA; and the
add-in software for data analysis), Statistical Package for the
Social Sciences, Minitab (State College, PA), and STATA
(StataCorp LP, College Station, TX) (EL Hajjar, 2011;
Gomez, 2010; Hamadu, Adeleke, & Ehie, 2011; Mills &
Raju, 2011; Spinelli, 2001). Few studies used some form of
computer managed learning tools such as The Learning Manager (WIN, Kingston, TN) (Cybinski & Selvanathan, 2005),
or STATLAB (Yale University, New Haven, CT) (Mills &
Raju, 2011). Other learning tools included Hawkes Learning
System, which was used at this university.
Hawkes Learning System. HLS is a mastery-based
software package, which is uploaded onto students’ personal
computers (Hawkes Learning Systems, 2012). The software
is an automatic homework and testing system and forms the
basis from which each student learns, practices, and masters
concepts within the topic being studied (HLS). The assignments are structured such that students must achieve an 80%
mastery level before they are certified in a particular topic
area. HLS provides computational problems, where students
must conduct required calculations outside of the program
(using pocket calculators, Microsoft Excel, or statistical software). The exams, which are actually created by the instructor
by selecting question types from a database of questions, will
188
G. R. SIMMONS
never be the same for any two students because the problems
within a question set are algorithmically generated.
HLS validity. HLS has not conducted formal investigations as to the validity and reliability of its exams. According
to R. Hendrix (personal communication, June 5, 2013), of
Hawkes Learning Systems, all questions used for either certifications or instructor developed exams are from the same
database of questions and have at least content validity.
time of the traditional 16-week term. Learning might be impacted by increased stress due to the intensive course schedule. Next, a delimitation of the study is that only exams were
used to measure progress and learning performance. Students
were required to only complete the HLS assignments and exams. The HLS assignments cannot be measured in that the
only grades identified if the student achieved 80% mastery.
Upon a student’s receipt of this mastery, the student would
receive full points for the assignment. Partial points were
granted based on lateness of submission of the assignment.
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Business Statistics Pedagogies
The School of Business teaches undergraduate business
statistics at the junior level. Students taking the course, regardless of teaching mode, receive instruction in descriptive
statistics and in the foundations of inferential statistics (Texas
A&M University-Central Texas, 2012). Each section of the
business statistics course is offered in one of three modes:
F2F, blended, and online. All modes use Blackboard 9.2
(Washington, DC) as the online component interface. Additionally, all modes use the HLS as the method of teaching
the foundations. Course length is either eight or 16 weeks.
F2F sections that are taught on main campus are 16 weeks.
Generally, all online sections are either in eight- or 16-week
sections, and all blended sections and any section taught in
the summer are eight weeks. The reason the blended courses
are eight weeks is because they are taught at the education
center on a local military base, where there is a contractual agreement that any course taught on base conforms to
an eight week course length. The F2F section receives traditional classroom lectures; the blended section downloads
and listens to recorded lectures and then attends questionand-answer sessions during class time; and the online section listens to recorded lectures and participates in discussion
threads.
The military base section is not a military-only section.
The course is offered on the course schedule with the other
sections and all students, regardless of military affiliation, are
free to register. Soldiers are not restricted to only taking the
military base section and may register for any section. Therefore, all sections, regardless of delivery mode and location,
have a homogeneous makeup of the university’s population.
The summer F2F sections may be taught either in blended
or traditional format, depending on the instructor’s desire.
Regardless of the length of the section, all students receive
the full course material as outlined in the master syllabi. Exams are administered to all students, regardless of teaching
mode or location, in the same manner, using the HLS. As the
instructor develops the exams by selecting question sets (not
individual questions), each exam tests the same concepts.
Study Limitations and Delimitation
A limitation and a delimitation of this study were identified.
First, the 8-week course length is a study limitation because
students are required to learn the course material in half the
Research Questions
In achieving the comparison of the performance in terms
of the three teaching modes, the following questions were
developed:
1. What was student performance within each teaching
mode in terms of course grades of the sections that
met between the fall of 2010 and the summer of 2012
at a regionally accredited university?
2. How did the performance of F2F, blended, and online
business statistics sections compare, in terms of course
exam scores, controlling for age, gender, experience
with statistics, and course length of the sections that
met between the fall of 2010 and the summer of 2012
at a regionally accredited university?
3. How did the performance by age group compare in
terms of the mean exam scores of the sections that met
between the fall of 2010 and the summer of 2012 at a
regionally accredited university?
The hypothesis that was tested was based on the second
and third research questions:
Hypothesis 1A (H1A ): There would be a difference in the
comparison of the performance of F2F, blended, and
online business statistics sections in terms of course
exam scores, controlling for age, gender, experience with
statistics, and course length of the sections that met between the fall of 2010 and the summer of 2012 at a
regionally accredited university.
H2A : There would be a difference in the comparison of
the performance by age group, in terms of the mean
exam scores, of the sections that met between the fall of
2010 and the summer of 2012 at a regionally accredited
university.
METHOD
Multivariate analysis of covariance (MANCOVA) was selected as the method to achieve the comparison of the modes.
MANCOVA was an appropriate method of analyzing the data
as the dependent variable performance was subdivided by the
three exams (Mertler & Vannatta, 2010). The remainder of
BUSINESS STATISTICS: A COMPARISON
this section describes the variables used, population and sampling, data collection, and analysis methodology.
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Variables
The dependent variable under test in this study was performance, in terms of the first three exams of each student in
each course section. The first exam score measured the performance of each student in understanding descriptive statistics and general probability principles. Descriptive statistics
included frequency distributions, measures of location (central tendency), and measures of dispersion (variance) as described in Hawkes and Marsh (2005). The second exam score
measured the performance of each student in understanding discrete distributions, including the binomial distribution,
sampling, and sampling distributions. The third exam score
measured the performance of each student in understanding
the estimation of means and proportions, hypothesis testing, and comparisons of population means and population
proportions. There was a fourth test, which measured the
understanding of three or more population comparisons and
the relationship of two or more continuous or categorical
variables; however, this course module was only recently
instituted and only a few of the course sections taught this
module; therefore, this module was not selected for inclusion
in the study. The independent variable was mode, which was
the teaching mode: F2F, blended, and online; gender, which
was the sex of each student (male or female); the covariates
were age; previous statistics experience (yes or no), which
was defined as an undergraduate 100- or 200-level statistics
course; and course length, which was either long (16 weeks)
or short (8 weeks). Grades were defined as the traditional
letter grades (A, B, C, D, and F) and were further defined as
A ≥ 90%, 80% ≤ B ≤ 89.99%, 70% ≤ C ≤ 79.99%, 60%
≤ D ≤ 69.99%, and F ≤ 60%. For the purpose of this study,
the course grades were calculated based on the mean score
of the three exams.
Population and Data Collection
The population in this study consisted of 13 total sections
of business statistics offered from the fall semester, in 2010
through the summer semester, in 2012. All sections were
taught by this researcher. Because all scores were available
to the researcher, and they could be easily manipulated in statistical software, all population units were used in the study.
Data that were excluded from this study were those from
students who had missing exam scores. Therefore, the total population, of exam scores (exams 1, 2, and 3) for all
students within the selected sections, was 440. The population was then divided into three subpopulations (or teaching
mode) named F2F (face to face or traditional), blended, and
online. The sizes of each of the subpopulations were 99,
122, and 219, respectively. All exam scores were collected
from the repository grade books in the instructor’s files. The
students that made up the population ranged in age from
189
20 to 62 years old. They were military members (soldiers,
airmen, or Marines), a Department of Defense employee, a
contractor for the Department of Defense, a family member
of the military (either spouse or child), or nonmilitary affiliated civilian. Additionally, some members of the military
were deployed overseas either in combative or noncombative
theaters of operation.
Analysis Methodology
The data collected from the grade books was cleaned and
entered into Minitab statistical software. Because the data included three response variables and a predictor variable, and
four covariates, a MANCOVA was selected as the statistical
method (Mertler & Vannatta, 2010; Ott & Longnecker, 2010).
All hypothesis tests were conducted with a significance level
of α = .05. The types of analysis included were the descriptive statistics for each sup-population (means, standard
deviations, and variances), mean effects plots, an ANOVA for
the differences found, including Bonferroni confidence intervals for comparing the significant differences as outlined in
Ott and Longnecker (2010).
ANALYSIS
The following analysis was conducted using Minitab
software and is presented below grouped into descriptive
statistics, MANCOVA tests, and comparison tests. All tests
were conducted according to the theories and concepts
presented by Mertler and Vannatta (2010) and Ott and
Longnecker (2010). These theories and concepts are also the
basis from which the tests could be analyzed in the Minitab
software.
Descriptive Statistics
As shown in Table 1, the means between each mode of exam
1 and exam 2 seem to be comparable; however, there seems to
be a difference in exam 3. The online exam 3 value is much
lower than either blended or F2F. The standard deviations
show some variability between each mode, within each exam.
In exam 1, online seemed to have the lowest variability and
F2F seemed to have the lowest variability in exams 2 and 3.
Figure 1 shows the variability within each exam. The
blended and online modes within exam 1 do show several
extreme outliers and mild outliers within the other exams. A
test for univariate normality revealed none of the exam scores
could be considered from a normal distribution (p < .005 for
each exam). Levene’s test for equal variances provided a p
value of .413 for exam 1; a p value of .079 for exam 2; and a p
value of .006 for exam 3; therefore, there was failure to reject
the equality of the variances in exams 1 and 2, but rejection
of the equality of variance in exam 3.
Figure 2 displays the scores by letter grade within each
mode by each exam. The line plots show that students
190
G. R. SIMMONS
TABLE 1
Descriptive Statistics: Performance by Teaching Mode
Variable
Mode
Total count
M
SD
Variance
IQR
Exam 1
Blended
F2F
Online
Blended
F2F
Online
Blended
F2F
Online
122
99
219
122
99
219
122
99
219
24.651
24.121
25.157
20.698
21.318
20.280
22.024
22.745
18.921
4.697
4.231
3.803
6.719
5.578
5.747
6.680
5.368
6.779
22.058
17.905
14.461
45.151
31.115
33.023
44.617
28.810
45.960
5.074
5.001
4.341
11.939
9.672
9.363
11.306
7.140
10.161
Exam 2
Exam 3
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Note: F2F = face to face; IQR = interquartile range.
performed better in exams 1 and 2, in terms of those scoring
Ds and Fs. In exam 3, the students scoring Ds and Fs scored
lower than in exams 1 and 2, with the students in the Online
mode scoring the lowest.
Table 2 shows the distribution of letter grades across the
different teaching modes.
retained as the only covariate in the model to the conduct of
the ANCOVA to determine the actual Bonferroni differences
within exam 3. The other covariates, gender, previous
statistics, and course length (p = .517, .244, and .970,
respectively), were found to be not significant.
Through the ANCOVA, in which age was controlled,
differences were found between the online mode and
both blended and F2F, but not between blended and F2F.
The Bonferroni confidence intervals were blended–online,
t = –4.630 p < .000 (–5.190, –1.643); and F2Fonline, t = –5.138 p < .000 (–5.903, –2.141). The
main effects plot in Figure 3 graphically depicts the
differences.
Multivariate Analysis of Covariance
In conducting the MANCOVA, performance (for exam 3)
was statistically significant when controlling for age, gender,
experience with statistics, and course length (Pillai’s Trace
= .108), F(6, 864) = 8.211, p < .000. Age (p = .015) was
Blended
Exam 1
F2F
Exam 2
30
30
25
20
20
15
10
10
0
Exam 3
30
20
10
0
Blended
F2F
Online
Mode
FIGURE 1
Boxplot of exam mean by teaching mode.
Online
BUSINESS STATISTICS: A COMPARISON
A
Exam 1
B
C
Exam 2
D
191
F
30
25
Mode
Blended
F2F
Online
20
Mean
15
10
Exam 3
30
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
25
20
15
10
A
B
C
D
F
Grade
FIGURE 2
Line plot of means (exams 1, 2, 3 vs. grade) (color figure available online).
Age
Age was also found to be statistically significant in the MANCOVA. Figure 4 is a histogram of the distribution of the ages
within the courses. As shown in the histogram, the distribution of ages is for a nontraditional university, where the
mean age is approximately 35 years (from Table 3). The ages
ranged from 20 to 62 years.
Table 3 shows the descriptive statistics for the age distribution. The age groups in Table 3 were based on the age
groups used by the National Center for Higher Education, in
their on-going research of higher education (National Center for Higher Education Management Systems Information
Center for Higher Education Policymaking and Analysis,
2013). The table shows the mean age of the students in these
courses was 34 years, which represents the largest age group
(25–34 years old). Approximately 54% of this age group took
the business statistics course online.
TABLE 2
Letter Grade Distribution by Teaching Mode
Grade
Blended
F2F
Online
A
B
C
D
F
Total
34%
16%
14%
10%
26%
100%
24%
27%
17%
12%
19%
100%
9%
16%
22%
11%
42%
100%
Note: F2F = face to face.
The 45–64 years old age group represented approximately
15% of the population. This age group was represented in
the three modes of instruction, blended (42%), F2F (27%),
and online (31%). Within each mode, the age group also had
the lowest numbers: blended (22%), F2F (17%), and online
(9%). It is important to note the 45–64 years old age group,
because this group had the lowest performance, in terms of
exam mean scores as shown in Figure 5.
Figure 5 shows that the 45–65 years old group had the
lowest average test score in each learning mode, except the
online mode, where the 25–34 years old group score highest
and the 45–65 years old group scored second highest, albeit
with low scores. In all year groups, except the 25–34 years
old group, students scored highest, within their year groups,
in the F2F learning mode.
In conducting a comparison of mean exam scores by age
group, a Kruskal-Wallis test was performed because the mean
exam score was determined to be nonnormal (p = .005).
A difference was found between the age groups using the
Kruskal-Wallis test, H(3) = 10.64, p = .014. In the actual
comparisons, the 25–34 years old age group was significantly
different from the 45–64 years old age group (p = .005) and
from the 35–44 years old age group (p = .029).
RESULTS
From the analysis (Table 1 and Figure 2), the blended and
F2F sections scored lowest on exam 2, on average, and had
an overall negative scoring trend, on average, as the sections
192
G. R. SIMMONS
23
21
20
19
Blended
F2F
Mode
FIGURE 3
Online
Main effects plot for exam 3.
12
Mean 34.61
8.562
SD
440
N
10
8
Percent
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Mean
22
6
4
2
0
16
24
FIGURE 4
32
40
Age
48
56
Age distribution of the population (color figure available online).
BUSINESS STATISTICS: A COMPARISON
193
TABLE 3
Age Distribution
Blended
Age group (years)
18–24
25–34
35–44
45–64
Total
F2F
Online
Count
M age (years)
SD
n
%
n
%
n
%
36
207
133
64
440
23.111
29.121
38.992
49.734
34.611
1.090
2.884
2.922
4.195
8.562
7
49
39
27
122
6
40
32
22
5
46
31
17
99
5
47
31
17
24
112
63
20
219
11
51
29
9
progressed through the course. The online sections had a continuous negative scoring trend, on average, as they progressed
through the course, scoring the lowest of each section, on average, on exam 3. The first research question asked about
student performance within each teaching mode in terms of
course grades. From Table 2, 50% of the students in the
blended sections earned As and Bs, whereas 36% earned Ds
and Fs. Within the F2F sections, 51% of the students earned
As and Bs, and 31% earned Ds and Fs. However, only 25%
of the students in the online sections earned As and Bs and
53% earned Ds and Fs.
The second research question asked how the performance
of F2F, blended, and online business statistics sections compared, in terms of course exam scores, controlling for age,
gender, experience with statistics, and course length. As
noted in the analysis section, statistical differences between
the blended and F2F sections were not found. However, the
linear combination between mode and exam 3, while controlling for age, was statistically significant. Additionally, the
significant differences were found between both the blended
and F2F modes and the online mode. Based on this difference, the null hypothesis was rejected.
The third research question asked how the performance
by age group compared in terms the mean exam scores.
In the analysis, the 35–44 and 45–64 years old age groups
were significantly different from the 25–34 years old age
group. The 18–24 years old age group was not significantly
different from any other age group. Based on the differences
from the 25–34 years old age group, the null hypothesis was
rejected.
Age Group
18–24
25–34
35–44
45–64
24
23
Exam Mean
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Note: F2F = face to face.
22
21
20
Blended
FIGURE 5
F2F
Mode
Online
Student grades by mode and age group (color figure available online).
194
G. R. SIMMONS
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
CONCLUSIONS
The purpose of this study was to compare the performance of
F2F, blended, and online business statistics sections in terms
of course exam scores, controlling for age, gender, previous
experience in statistics, and course length, of the sections
that met between the fall of 2010 and the summer of 2012
at a regionally accredited university. The results found in the
blended and F2F teaching modes that nearly half of each
sub-population scored As or Bs, on average, of the exams,
but only 25% of the online students scored As or Bs, on
average. A significant difference was found in the online
teaching mode for exam 3 leading to the rejection of the null
hypothesis and a significant difference was found between
the age groups.
Student performance in terms of course grades in the first
research question showed students performed better in the
both the blended and F2F classes, than in the online classes.
The reasons for the differences between the modes was possibly due to the online students not viewing the recorded
lectures prior to beginning practice problems or the homework assignments. The single common factor between the
F2F and the blended was direct or physical access to the
instructor during class time.
Student performance in terms of the three teaching modes
and exam scores, in the second research question, showed no
significant difference between the F2F and blended modes,
but did show a significant difference between those modes
and the online mode in the linear combination of the third
exam. This led to the rejection of the first hypothesis. There
was a trend of lower tests scores as the course progressed. The
differences in the third exam for the online students, seems
to be a combination of the difficulty and unfamiliarity of the
concepts and possibly the exam was taken at the end of the
term, where other personal factors or stresses, not considered
in this study, may have affected the students.
Student performance in terms of age groups and mean
exam scores, in the third research question, showed differences found with age groups, the 45–64 years old age group
consistently performed lower than the other age groups in
blended and F2F. This difference led to the rejection of the
second hypothesis. The differences in the 45–64 years old
age group might possibly be the result of the span of time
because they were last in an academic learning environment.
Students in this age may need to consider taking or retaking math or other quantitative courses as a refresher prior to
taking the business statistics course.
RECOMMENDATIONS FOR FUTURE STUDIES
This study has surfaced problem areas in learning business
statistics, at this institution, primarily for the online students.
Further research is required to determine the root causes of
these problems. At the time of the writing, the institution
identified courses, including business statistics, that would
only be taught over 16 weeks in the online mode. This rule
was be implemented in the fall 2013 term. There may be
additional factors or student stressors that were not considered, but should be identified. Future studies should identify
these additional factors, perhaps time management, procrastination, or test anxiety on the part of the students, and include them in the analysis. Finally, identify levels of previous
mathematical or other quantitative skills, by age group, to determine the causes of the differences in exams scores.
REFERENCES
Arbaugh, J. B., Desai, A., Rau, B., & Sridhar, B. S. (2010). A review
of research on online and blended learning in the management disciplines:
1994–2009. Organization Management Journal, 7, 39–55. http://dx.doi.
org/10.1057/omj.2010.5
Ashby, J., Sadera, W. A., & McNary, S. W. (2011). Comparing student
success between development math courses offered online, blended, and
face-to-face. Journal of Interactive Online Learning, 10, 129–140. Retrieved from ProQuest database.
Bernard, R. M., Abrami, P. C., Lou, Y., Borokhovski, E., Wade, A., Wozney,
L., . . .Huang, B. (2004). How does distance education compare with
classroom instruction? A meta-analysis of the empirical literature. Review of Educational Research, 74, 379–439. Retrieved from ProQuest
database.
Cybinski, P., & Selvanathan, S. (2005). Learning experience and learning
effectiveness in undergraduate statistics: Modeling performance in traditional and flexible learning environments. Decision Sciences Journal
of Innovative Education, 3, 251–271. http://dx.doi.org/10.1111/j.15404609.2005.00069.x
DeNeui, D. L., & Dodge, T. L. (2006). Asynchronous learning networks
and student outcomes: the utility of online learning components in hybrid
courses. Journal of Instructional Psychology, 33, 256–259. Retrieved
from ProQuest database.
Dotterweich, D. P., & Rochelle, C. F. (2012, Mar/Apr). Online, instructional television, and traditional delivery: Student characteristics and
success factors in business statistics. American Journal of Business, 5,
129–138. Retrieved from http://ehis.ebscohost.com.zeus.tarleton.edu:81/
ehost/pdfviewer/pdfviewer?vid=9&hid=3&sid=965a771a-4662-47b08080-8da49b6f4146%40sessionmgr112
El Hajjar, S. T. (2011). An empirical study about the use and implementation of software in statistics at higher education institutions. International
Journal of Engineering & Technology, 11(6), 45–54. Retrieved from ProQuest database.
Gomez, R. (2010). Innovations in teaching undergraduate statistics courses
for biology students. Review of Higher Education and Self-Learning, 3(7),
8–13. Retrieved from ProQuest database.
Hamadu, D., Adeleke, I., & Ehie, I. (2011). Using information technology
in teaching of business statistics in Nigeria Business School. American
Journal of Business Education, 4(10), 85–92. Retrieved from ProQuest
database.
Harrington, D. (1999). Teaching Statistics: A comparison of traditional
classroom and programmed instruction/distance learning approaches.
Journal of Social Work Education, 35, 343–352. Retrieved from ProQuest
database.
Hawkes, J. S., & Marsh, W. H. (2005). Discovering statistics (2nd ed.).
Charleston, SC: Hawkes Learning Systems.
Hawkes Learning Systems. (2012). How the software works. Retrieved from
http://www.hawkeslearning.com/Instructors/HowTheSoftwareWorks.htm
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
BUSINESS STATISTICS: A COMPARISON
McLaren, C. H. (2004). A comparison of student persistence and performance in online and classroom business statistics experiences. Decision Sciences Journal of Innovative Education, 2, 1–10. http://dx.doi.org/
10.1111/j.0011-7315.2004.00015.x.
Mertler, C. A., & Vannatta, R. A. (2010). Advanced and multivariate statistical methods (4th ed.). Glendale, CA: Pyrczak.
Mills, J. D., & Raju, D. (2011). Teaching statistics online: A decade’s
review of the literature about what works. Journal of Statistics Education, 19, 1–27. Retrieved from http://www.amstat.org/publications/
jse/v19n2/mills.pdf
National Center for Education Statistics. (2011). Fast facts. Retrieved from
http://nces.ed.gov/fastfacts/display.asp?id=74
National Center for Higher Education Management Systems Information
Center for Higher Education Policymaking and Analysis. (2013). Educational levels of the population: Educational attainment by degree-level
and age-group (decennial census). Retrieved from http://www.highered
info.org/dbrowser/index.php?submeasure=201&year=2000&level=nat
ion&mode=data&state=0
Ott, R. L., & Longnecker, M. (2010). An introduction to statistical methods
and data analysis (6th ed.). Belmont, CA: Brooks/Cole–Cengage.
Radford, A. W. (2011, October). Learning at a distance: undergraduate enrollment in distance education courses and degree programs.
195
NCES 2012-154. Stats in Brief. Retrieved from http://nces.ed.gov/pubs
2012/2012154.pdf
Spinelli, M. A. (2001). The use of technology in teaching business statistics.
Journal of Education for Business, 77, 41–43. Retrieved from ProQuest
database.
Summers, J. J., Waigandt, A., & Whittaker, T. A. (2005). A comparison
of student achievement and satisfaction in an online versus a traditional
face-to-face statistics class. Innovative Higher Education, 29, 233–250.
http://dx.doi.org/10.1007/s10755-005-1938-x
Tallent-Runnels, M. K., Thomas, J. A., Lan, W. Y., Cooper, S., Ahern, T.
C., Shaw, S. M., & Liu, X. (2006). Teaching courses online: A review
of the research. Review of Educational Research, 76, 93–135. Retrieved
from ProQuest database.
Texas A&M University-Central Texas. (2012). Academic class
syllabi. Retrieved from http://www.tamuct.edu/departments/syllabi/
index.php
University of Phoenix. (2012). History. Retrieved from http://www.
phoenix.edu/about us/about university of phoenix/history.html
York, R. O. (2008). Comparing three modes of instruction in a graduate social work program. Journal of Social Work Education, 44, 157–172. Retrieved from ProQuest
database.
ISSN: 0883-2323 (Print) 1940-3356 (Online) Journal homepage: http://www.tandfonline.com/loi/vjeb20
Business Statistics: A Comparison of Student
Performance in Three Learning Modes
Gerald R. Simmons
To cite this article: Gerald R. Simmons (2014) Business Statistics: A Comparison of Student
Performance in Three Learning Modes, Journal of Education for Business, 89:4, 186-195, DOI:
10.1080/08832323.2013.836470
To link to this article: http://dx.doi.org/10.1080/08832323.2013.836470
Published online: 02 May 2014.
Submit your article to this journal
Article views: 140
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=vjeb20
Download by: [Universitas Maritim Raja Ali Haji]
Date: 11 January 2016, At: 20:39
JOURNAL OF EDUCATION FOR BUSINESS, 89: 186–195, 2014
C Taylor & Francis Group, LLC
Copyright
ISSN: 0883-2323 print / 1940-3356 online
DOI: 10.1080/08832323.2013.836470
Business Statistics: A Comparison of Student
Performance in Three Learning Modes
Gerald R. Simmons
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Texas A&M University-Central Texas, Killeen, Texas, USA
The purpose of this study was to compare the performance of three teaching modes and age
groups of business statistics sections in terms of course exam scores. The research questions
were formulated to determine the performance of the students within each teaching mode, to
compare each mode in terms of exam scores, and to compare exam scores by age group. The
research hypotheses predicted there would be a difference between the three teaching modes
and age groups. The results found significant differences in both the online teaching mode for
exam 3 and age groups for the mean of the three exams.
Keywords: blended, business statistics, face-to-face, Hawkes Learning Systems, online, teaching modes
As we fast approach the closing of the first quarter of this century, methods of receiving higher level education continues
to change from the traditional mode of university learning
in face-to-face (F2F) lectures and labs. Online learning is
quickly becoming the new traditional method of learning.
Radford (2011) found that between 2000 and 2008 the percentages of undergraduates enrolled in at least one distance
education class expanded from 8% to 20%. Additionally,
the percentage of undergraduates enrolling in distance education programs increased from two to four percent during
the same period (Radford, 2011). Of all students, Radford
also found 24% of the business students enrolled in distance
education courses and 6% enrolled in distance education
programs. In terms of enrollment, the 2009 figures show
University of Phoenix, Online Campus, has the highest enrollment of all U.S. postsecondary institutions, with 380,232
students (National Center for Education Statistics, 2011).
Kaplan University is second, with 71,011 students, and Arizona State University is third with 68,064 students (National
Center for Education Statistics, 2011). The other two institutions that made the top five were Miami Dade College
at 59,120 students and Ohio State University, Main Campus, at 55,014 students. The University of Minnesota, Twin
Correspondence should be addressed to Gerald R. Simmons, Texas
A&M University-Central Texas, Department of Management and Marketing, 1001 Leadership Place, Founders’ Hall, Killeen, TX 76549, USA.
E-mail: [email protected]
Cities, was rated ninth; the University of Texas at Austin
was rated 10th; Texas A&M University was 12th; and the
University of Washington, Seattle Campus was rated 18th
(National Center for Education Statistics, 2011). What is interesting to note is the fact the top two institutions are fully
online universities. Additionally, most of the listed traditional
universities either have fully online or fully distance degree
programs.
This regionally accredited university provides it students
with upper level (baccalaureate) and graduate programs and
provides both online, blended, and traditional courses and
degree programs. The objective of this study was to expand
the current knowledge concerning the perceptions of online
learning versus the perceptions of traditional learning; therefore, this study presents a comparison of online, blended, and
traditional F2F business statistics sections at a single university. The purpose of this quantitative study was to compare the
performance of F2F, blended, and online business statistics
sections in terms of course exam scores, controlling for age,
gender, previous experience in statistics, and course length,
of the sections that met between the fall of 2010 and the summer of 2012 at a regionally accredited university. This study
was differentiated from other online, F2F, and blended comparison studies in that the data collected were from sections
that were taught using the same textbooks, as in the studies
by Dotterweich and Rochelle (2012) and McLaren (2004),
but also administered the same assignments and exams in the
same manner; thus reducing possible instructor bias. These
details will be explained in the following discussion of the
Hawkes Learning System (HLS)
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
BUSINESS STATISTICS: A COMPARISON
This study differed from Dotterweich and Rochelle whose
online students completed online assignments while the F2F
students completed the traditional assignments from the textbook. This study differed from McLaren (2004), whose F2F
and online assignments were also presented in a different
form and the exams were administered differently. F2F students could take their exams with a sheet of notes, whereas
the online students had access to all notes and textbooks, with
the request they do not receive outside assistance (McLaren,
2004). Additionally, this study sought to determine the effects of the course length. The university taught the courses
in eight and sixteen week sessions. The following sections
provide a brief background of online learning, the online
and F2F pedagogies of the regionally accredited university’s
business statistics courses, methodology, analysis, and finally
the results of the study.
BACKGROUND
Online Learning
The roots of distance education and online learning stemmed
from correspondence courses. These courses were taught
over a geographic distance where instructors and students
corresponded through the postal system. With the introduction of the Internet and the world wide web, in 1989, online
and distance education began its expansive growth. It was
at this time that institutions, such as University of Phoenix,
established their online programs (University of Phoenix,
2012). The advantage of the Internet and world wide web
was the quick transmission and correspondence between the
instructors and the students. Online learning grew from email
transmissions, to newsgroups, to the online learning management systems (e.g., Blackboard, Moodle, Desire2Learn).
Online Versus Blended Versus Traditional
Learning
Online learning consists of students receiving instruction,
conducting research, completing assignments, quizzes, and
exams, and interacting with the instructor in a format that uses
the world wide web as the basis for a learning management
system. This interaction can be either in a synchronous mode,
or in an asynchronous mode (Mills & Raju, 2011; Summers,
Waigandt, & Whittaker, 2005; Tallent-Runnels et al., 2006).
Blended learning courses consists of traditional F2F lecture
periods, coupled with online components such as group or
team study, assignments, quizzes, and exams (Mills & Raju,
2011). Traditional learning consists of the F2F interactions
according to scheduled periods, where students receive information (or lectures) in a classroom.
The idea that traditional learners perform better than online learners has pervaded the thoughts of professionals and
academicians since the advent of distance learning. This idea
became more prevalent with the growing numbers of ac-
187
credited online universities and their growing populations of
students and the online programs within traditional universities, spurring many studies comparing the effects of online
learning and traditional learning (DeNeui & Dodge, 2006;
Harrington, 1999; Mills & Raju, 2011; Summers et al., 2005;
York, 2008). Studies of comparison sought to determine if
there were true differences in student learning between online
and F2F (Bernard et al., 2004; Harrington, 1999; Summers
et al., 2005), blended and online (Arbaugh, Desai, Rau, &
Sridhar, 2010), and all three (Ashby, Sadera, & McNary,
2011; York, 2008). The results of these studies included a
poorer performance in the traditional classroom (Ashby et al.,
2011) as well as no significant difference in the three modes
of instruction (Harrington, 1999; Summers et al., 2005; York,
2008). All researchers recommended further study, especially
with the expected increase of online learning.
Business Statistics and Hawkes Learning
System
Technology has always taken a role in the education of students, beginning with the abacus and continuing on to the
laptop and notebook computers of today. Without the use of
technology, courses such as statistics became the dread of
many students whose intent was to learn only the basics, not
actually major in the subject.
Technologies used in many introductory statistics courses,
regardless of teaching mode primarily consisted of some
form of computational software. These software packages
included the Texas Instruments 84 (Dallas, TX) family of
pocket calculators, Microsoft Excel (Seattle, WA; and the
add-in software for data analysis), Statistical Package for the
Social Sciences, Minitab (State College, PA), and STATA
(StataCorp LP, College Station, TX) (EL Hajjar, 2011;
Gomez, 2010; Hamadu, Adeleke, & Ehie, 2011; Mills &
Raju, 2011; Spinelli, 2001). Few studies used some form of
computer managed learning tools such as The Learning Manager (WIN, Kingston, TN) (Cybinski & Selvanathan, 2005),
or STATLAB (Yale University, New Haven, CT) (Mills &
Raju, 2011). Other learning tools included Hawkes Learning
System, which was used at this university.
Hawkes Learning System. HLS is a mastery-based
software package, which is uploaded onto students’ personal
computers (Hawkes Learning Systems, 2012). The software
is an automatic homework and testing system and forms the
basis from which each student learns, practices, and masters
concepts within the topic being studied (HLS). The assignments are structured such that students must achieve an 80%
mastery level before they are certified in a particular topic
area. HLS provides computational problems, where students
must conduct required calculations outside of the program
(using pocket calculators, Microsoft Excel, or statistical software). The exams, which are actually created by the instructor
by selecting question types from a database of questions, will
188
G. R. SIMMONS
never be the same for any two students because the problems
within a question set are algorithmically generated.
HLS validity. HLS has not conducted formal investigations as to the validity and reliability of its exams. According
to R. Hendrix (personal communication, June 5, 2013), of
Hawkes Learning Systems, all questions used for either certifications or instructor developed exams are from the same
database of questions and have at least content validity.
time of the traditional 16-week term. Learning might be impacted by increased stress due to the intensive course schedule. Next, a delimitation of the study is that only exams were
used to measure progress and learning performance. Students
were required to only complete the HLS assignments and exams. The HLS assignments cannot be measured in that the
only grades identified if the student achieved 80% mastery.
Upon a student’s receipt of this mastery, the student would
receive full points for the assignment. Partial points were
granted based on lateness of submission of the assignment.
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Business Statistics Pedagogies
The School of Business teaches undergraduate business
statistics at the junior level. Students taking the course, regardless of teaching mode, receive instruction in descriptive
statistics and in the foundations of inferential statistics (Texas
A&M University-Central Texas, 2012). Each section of the
business statistics course is offered in one of three modes:
F2F, blended, and online. All modes use Blackboard 9.2
(Washington, DC) as the online component interface. Additionally, all modes use the HLS as the method of teaching
the foundations. Course length is either eight or 16 weeks.
F2F sections that are taught on main campus are 16 weeks.
Generally, all online sections are either in eight- or 16-week
sections, and all blended sections and any section taught in
the summer are eight weeks. The reason the blended courses
are eight weeks is because they are taught at the education
center on a local military base, where there is a contractual agreement that any course taught on base conforms to
an eight week course length. The F2F section receives traditional classroom lectures; the blended section downloads
and listens to recorded lectures and then attends questionand-answer sessions during class time; and the online section listens to recorded lectures and participates in discussion
threads.
The military base section is not a military-only section.
The course is offered on the course schedule with the other
sections and all students, regardless of military affiliation, are
free to register. Soldiers are not restricted to only taking the
military base section and may register for any section. Therefore, all sections, regardless of delivery mode and location,
have a homogeneous makeup of the university’s population.
The summer F2F sections may be taught either in blended
or traditional format, depending on the instructor’s desire.
Regardless of the length of the section, all students receive
the full course material as outlined in the master syllabi. Exams are administered to all students, regardless of teaching
mode or location, in the same manner, using the HLS. As the
instructor develops the exams by selecting question sets (not
individual questions), each exam tests the same concepts.
Study Limitations and Delimitation
A limitation and a delimitation of this study were identified.
First, the 8-week course length is a study limitation because
students are required to learn the course material in half the
Research Questions
In achieving the comparison of the performance in terms
of the three teaching modes, the following questions were
developed:
1. What was student performance within each teaching
mode in terms of course grades of the sections that
met between the fall of 2010 and the summer of 2012
at a regionally accredited university?
2. How did the performance of F2F, blended, and online
business statistics sections compare, in terms of course
exam scores, controlling for age, gender, experience
with statistics, and course length of the sections that
met between the fall of 2010 and the summer of 2012
at a regionally accredited university?
3. How did the performance by age group compare in
terms of the mean exam scores of the sections that met
between the fall of 2010 and the summer of 2012 at a
regionally accredited university?
The hypothesis that was tested was based on the second
and third research questions:
Hypothesis 1A (H1A ): There would be a difference in the
comparison of the performance of F2F, blended, and
online business statistics sections in terms of course
exam scores, controlling for age, gender, experience with
statistics, and course length of the sections that met between the fall of 2010 and the summer of 2012 at a
regionally accredited university.
H2A : There would be a difference in the comparison of
the performance by age group, in terms of the mean
exam scores, of the sections that met between the fall of
2010 and the summer of 2012 at a regionally accredited
university.
METHOD
Multivariate analysis of covariance (MANCOVA) was selected as the method to achieve the comparison of the modes.
MANCOVA was an appropriate method of analyzing the data
as the dependent variable performance was subdivided by the
three exams (Mertler & Vannatta, 2010). The remainder of
BUSINESS STATISTICS: A COMPARISON
this section describes the variables used, population and sampling, data collection, and analysis methodology.
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Variables
The dependent variable under test in this study was performance, in terms of the first three exams of each student in
each course section. The first exam score measured the performance of each student in understanding descriptive statistics and general probability principles. Descriptive statistics
included frequency distributions, measures of location (central tendency), and measures of dispersion (variance) as described in Hawkes and Marsh (2005). The second exam score
measured the performance of each student in understanding discrete distributions, including the binomial distribution,
sampling, and sampling distributions. The third exam score
measured the performance of each student in understanding
the estimation of means and proportions, hypothesis testing, and comparisons of population means and population
proportions. There was a fourth test, which measured the
understanding of three or more population comparisons and
the relationship of two or more continuous or categorical
variables; however, this course module was only recently
instituted and only a few of the course sections taught this
module; therefore, this module was not selected for inclusion
in the study. The independent variable was mode, which was
the teaching mode: F2F, blended, and online; gender, which
was the sex of each student (male or female); the covariates
were age; previous statistics experience (yes or no), which
was defined as an undergraduate 100- or 200-level statistics
course; and course length, which was either long (16 weeks)
or short (8 weeks). Grades were defined as the traditional
letter grades (A, B, C, D, and F) and were further defined as
A ≥ 90%, 80% ≤ B ≤ 89.99%, 70% ≤ C ≤ 79.99%, 60%
≤ D ≤ 69.99%, and F ≤ 60%. For the purpose of this study,
the course grades were calculated based on the mean score
of the three exams.
Population and Data Collection
The population in this study consisted of 13 total sections
of business statistics offered from the fall semester, in 2010
through the summer semester, in 2012. All sections were
taught by this researcher. Because all scores were available
to the researcher, and they could be easily manipulated in statistical software, all population units were used in the study.
Data that were excluded from this study were those from
students who had missing exam scores. Therefore, the total population, of exam scores (exams 1, 2, and 3) for all
students within the selected sections, was 440. The population was then divided into three subpopulations (or teaching
mode) named F2F (face to face or traditional), blended, and
online. The sizes of each of the subpopulations were 99,
122, and 219, respectively. All exam scores were collected
from the repository grade books in the instructor’s files. The
students that made up the population ranged in age from
189
20 to 62 years old. They were military members (soldiers,
airmen, or Marines), a Department of Defense employee, a
contractor for the Department of Defense, a family member
of the military (either spouse or child), or nonmilitary affiliated civilian. Additionally, some members of the military
were deployed overseas either in combative or noncombative
theaters of operation.
Analysis Methodology
The data collected from the grade books was cleaned and
entered into Minitab statistical software. Because the data included three response variables and a predictor variable, and
four covariates, a MANCOVA was selected as the statistical
method (Mertler & Vannatta, 2010; Ott & Longnecker, 2010).
All hypothesis tests were conducted with a significance level
of α = .05. The types of analysis included were the descriptive statistics for each sup-population (means, standard
deviations, and variances), mean effects plots, an ANOVA for
the differences found, including Bonferroni confidence intervals for comparing the significant differences as outlined in
Ott and Longnecker (2010).
ANALYSIS
The following analysis was conducted using Minitab
software and is presented below grouped into descriptive
statistics, MANCOVA tests, and comparison tests. All tests
were conducted according to the theories and concepts
presented by Mertler and Vannatta (2010) and Ott and
Longnecker (2010). These theories and concepts are also the
basis from which the tests could be analyzed in the Minitab
software.
Descriptive Statistics
As shown in Table 1, the means between each mode of exam
1 and exam 2 seem to be comparable; however, there seems to
be a difference in exam 3. The online exam 3 value is much
lower than either blended or F2F. The standard deviations
show some variability between each mode, within each exam.
In exam 1, online seemed to have the lowest variability and
F2F seemed to have the lowest variability in exams 2 and 3.
Figure 1 shows the variability within each exam. The
blended and online modes within exam 1 do show several
extreme outliers and mild outliers within the other exams. A
test for univariate normality revealed none of the exam scores
could be considered from a normal distribution (p < .005 for
each exam). Levene’s test for equal variances provided a p
value of .413 for exam 1; a p value of .079 for exam 2; and a p
value of .006 for exam 3; therefore, there was failure to reject
the equality of the variances in exams 1 and 2, but rejection
of the equality of variance in exam 3.
Figure 2 displays the scores by letter grade within each
mode by each exam. The line plots show that students
190
G. R. SIMMONS
TABLE 1
Descriptive Statistics: Performance by Teaching Mode
Variable
Mode
Total count
M
SD
Variance
IQR
Exam 1
Blended
F2F
Online
Blended
F2F
Online
Blended
F2F
Online
122
99
219
122
99
219
122
99
219
24.651
24.121
25.157
20.698
21.318
20.280
22.024
22.745
18.921
4.697
4.231
3.803
6.719
5.578
5.747
6.680
5.368
6.779
22.058
17.905
14.461
45.151
31.115
33.023
44.617
28.810
45.960
5.074
5.001
4.341
11.939
9.672
9.363
11.306
7.140
10.161
Exam 2
Exam 3
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Note: F2F = face to face; IQR = interquartile range.
performed better in exams 1 and 2, in terms of those scoring
Ds and Fs. In exam 3, the students scoring Ds and Fs scored
lower than in exams 1 and 2, with the students in the Online
mode scoring the lowest.
Table 2 shows the distribution of letter grades across the
different teaching modes.
retained as the only covariate in the model to the conduct of
the ANCOVA to determine the actual Bonferroni differences
within exam 3. The other covariates, gender, previous
statistics, and course length (p = .517, .244, and .970,
respectively), were found to be not significant.
Through the ANCOVA, in which age was controlled,
differences were found between the online mode and
both blended and F2F, but not between blended and F2F.
The Bonferroni confidence intervals were blended–online,
t = –4.630 p < .000 (–5.190, –1.643); and F2Fonline, t = –5.138 p < .000 (–5.903, –2.141). The
main effects plot in Figure 3 graphically depicts the
differences.
Multivariate Analysis of Covariance
In conducting the MANCOVA, performance (for exam 3)
was statistically significant when controlling for age, gender,
experience with statistics, and course length (Pillai’s Trace
= .108), F(6, 864) = 8.211, p < .000. Age (p = .015) was
Blended
Exam 1
F2F
Exam 2
30
30
25
20
20
15
10
10
0
Exam 3
30
20
10
0
Blended
F2F
Online
Mode
FIGURE 1
Boxplot of exam mean by teaching mode.
Online
BUSINESS STATISTICS: A COMPARISON
A
Exam 1
B
C
Exam 2
D
191
F
30
25
Mode
Blended
F2F
Online
20
Mean
15
10
Exam 3
30
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
25
20
15
10
A
B
C
D
F
Grade
FIGURE 2
Line plot of means (exams 1, 2, 3 vs. grade) (color figure available online).
Age
Age was also found to be statistically significant in the MANCOVA. Figure 4 is a histogram of the distribution of the ages
within the courses. As shown in the histogram, the distribution of ages is for a nontraditional university, where the
mean age is approximately 35 years (from Table 3). The ages
ranged from 20 to 62 years.
Table 3 shows the descriptive statistics for the age distribution. The age groups in Table 3 were based on the age
groups used by the National Center for Higher Education, in
their on-going research of higher education (National Center for Higher Education Management Systems Information
Center for Higher Education Policymaking and Analysis,
2013). The table shows the mean age of the students in these
courses was 34 years, which represents the largest age group
(25–34 years old). Approximately 54% of this age group took
the business statistics course online.
TABLE 2
Letter Grade Distribution by Teaching Mode
Grade
Blended
F2F
Online
A
B
C
D
F
Total
34%
16%
14%
10%
26%
100%
24%
27%
17%
12%
19%
100%
9%
16%
22%
11%
42%
100%
Note: F2F = face to face.
The 45–64 years old age group represented approximately
15% of the population. This age group was represented in
the three modes of instruction, blended (42%), F2F (27%),
and online (31%). Within each mode, the age group also had
the lowest numbers: blended (22%), F2F (17%), and online
(9%). It is important to note the 45–64 years old age group,
because this group had the lowest performance, in terms of
exam mean scores as shown in Figure 5.
Figure 5 shows that the 45–65 years old group had the
lowest average test score in each learning mode, except the
online mode, where the 25–34 years old group score highest
and the 45–65 years old group scored second highest, albeit
with low scores. In all year groups, except the 25–34 years
old group, students scored highest, within their year groups,
in the F2F learning mode.
In conducting a comparison of mean exam scores by age
group, a Kruskal-Wallis test was performed because the mean
exam score was determined to be nonnormal (p = .005).
A difference was found between the age groups using the
Kruskal-Wallis test, H(3) = 10.64, p = .014. In the actual
comparisons, the 25–34 years old age group was significantly
different from the 45–64 years old age group (p = .005) and
from the 35–44 years old age group (p = .029).
RESULTS
From the analysis (Table 1 and Figure 2), the blended and
F2F sections scored lowest on exam 2, on average, and had
an overall negative scoring trend, on average, as the sections
192
G. R. SIMMONS
23
21
20
19
Blended
F2F
Mode
FIGURE 3
Online
Main effects plot for exam 3.
12
Mean 34.61
8.562
SD
440
N
10
8
Percent
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Mean
22
6
4
2
0
16
24
FIGURE 4
32
40
Age
48
56
Age distribution of the population (color figure available online).
BUSINESS STATISTICS: A COMPARISON
193
TABLE 3
Age Distribution
Blended
Age group (years)
18–24
25–34
35–44
45–64
Total
F2F
Online
Count
M age (years)
SD
n
%
n
%
n
%
36
207
133
64
440
23.111
29.121
38.992
49.734
34.611
1.090
2.884
2.922
4.195
8.562
7
49
39
27
122
6
40
32
22
5
46
31
17
99
5
47
31
17
24
112
63
20
219
11
51
29
9
progressed through the course. The online sections had a continuous negative scoring trend, on average, as they progressed
through the course, scoring the lowest of each section, on average, on exam 3. The first research question asked about
student performance within each teaching mode in terms of
course grades. From Table 2, 50% of the students in the
blended sections earned As and Bs, whereas 36% earned Ds
and Fs. Within the F2F sections, 51% of the students earned
As and Bs, and 31% earned Ds and Fs. However, only 25%
of the students in the online sections earned As and Bs and
53% earned Ds and Fs.
The second research question asked how the performance
of F2F, blended, and online business statistics sections compared, in terms of course exam scores, controlling for age,
gender, experience with statistics, and course length. As
noted in the analysis section, statistical differences between
the blended and F2F sections were not found. However, the
linear combination between mode and exam 3, while controlling for age, was statistically significant. Additionally, the
significant differences were found between both the blended
and F2F modes and the online mode. Based on this difference, the null hypothesis was rejected.
The third research question asked how the performance
by age group compared in terms the mean exam scores.
In the analysis, the 35–44 and 45–64 years old age groups
were significantly different from the 25–34 years old age
group. The 18–24 years old age group was not significantly
different from any other age group. Based on the differences
from the 25–34 years old age group, the null hypothesis was
rejected.
Age Group
18–24
25–34
35–44
45–64
24
23
Exam Mean
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
Note: F2F = face to face.
22
21
20
Blended
FIGURE 5
F2F
Mode
Online
Student grades by mode and age group (color figure available online).
194
G. R. SIMMONS
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
CONCLUSIONS
The purpose of this study was to compare the performance of
F2F, blended, and online business statistics sections in terms
of course exam scores, controlling for age, gender, previous
experience in statistics, and course length, of the sections
that met between the fall of 2010 and the summer of 2012
at a regionally accredited university. The results found in the
blended and F2F teaching modes that nearly half of each
sub-population scored As or Bs, on average, of the exams,
but only 25% of the online students scored As or Bs, on
average. A significant difference was found in the online
teaching mode for exam 3 leading to the rejection of the null
hypothesis and a significant difference was found between
the age groups.
Student performance in terms of course grades in the first
research question showed students performed better in the
both the blended and F2F classes, than in the online classes.
The reasons for the differences between the modes was possibly due to the online students not viewing the recorded
lectures prior to beginning practice problems or the homework assignments. The single common factor between the
F2F and the blended was direct or physical access to the
instructor during class time.
Student performance in terms of the three teaching modes
and exam scores, in the second research question, showed no
significant difference between the F2F and blended modes,
but did show a significant difference between those modes
and the online mode in the linear combination of the third
exam. This led to the rejection of the first hypothesis. There
was a trend of lower tests scores as the course progressed. The
differences in the third exam for the online students, seems
to be a combination of the difficulty and unfamiliarity of the
concepts and possibly the exam was taken at the end of the
term, where other personal factors or stresses, not considered
in this study, may have affected the students.
Student performance in terms of age groups and mean
exam scores, in the third research question, showed differences found with age groups, the 45–64 years old age group
consistently performed lower than the other age groups in
blended and F2F. This difference led to the rejection of the
second hypothesis. The differences in the 45–64 years old
age group might possibly be the result of the span of time
because they were last in an academic learning environment.
Students in this age may need to consider taking or retaking math or other quantitative courses as a refresher prior to
taking the business statistics course.
RECOMMENDATIONS FOR FUTURE STUDIES
This study has surfaced problem areas in learning business
statistics, at this institution, primarily for the online students.
Further research is required to determine the root causes of
these problems. At the time of the writing, the institution
identified courses, including business statistics, that would
only be taught over 16 weeks in the online mode. This rule
was be implemented in the fall 2013 term. There may be
additional factors or student stressors that were not considered, but should be identified. Future studies should identify
these additional factors, perhaps time management, procrastination, or test anxiety on the part of the students, and include them in the analysis. Finally, identify levels of previous
mathematical or other quantitative skills, by age group, to determine the causes of the differences in exams scores.
REFERENCES
Arbaugh, J. B., Desai, A., Rau, B., & Sridhar, B. S. (2010). A review
of research on online and blended learning in the management disciplines:
1994–2009. Organization Management Journal, 7, 39–55. http://dx.doi.
org/10.1057/omj.2010.5
Ashby, J., Sadera, W. A., & McNary, S. W. (2011). Comparing student
success between development math courses offered online, blended, and
face-to-face. Journal of Interactive Online Learning, 10, 129–140. Retrieved from ProQuest database.
Bernard, R. M., Abrami, P. C., Lou, Y., Borokhovski, E., Wade, A., Wozney,
L., . . .Huang, B. (2004). How does distance education compare with
classroom instruction? A meta-analysis of the empirical literature. Review of Educational Research, 74, 379–439. Retrieved from ProQuest
database.
Cybinski, P., & Selvanathan, S. (2005). Learning experience and learning
effectiveness in undergraduate statistics: Modeling performance in traditional and flexible learning environments. Decision Sciences Journal
of Innovative Education, 3, 251–271. http://dx.doi.org/10.1111/j.15404609.2005.00069.x
DeNeui, D. L., & Dodge, T. L. (2006). Asynchronous learning networks
and student outcomes: the utility of online learning components in hybrid
courses. Journal of Instructional Psychology, 33, 256–259. Retrieved
from ProQuest database.
Dotterweich, D. P., & Rochelle, C. F. (2012, Mar/Apr). Online, instructional television, and traditional delivery: Student characteristics and
success factors in business statistics. American Journal of Business, 5,
129–138. Retrieved from http://ehis.ebscohost.com.zeus.tarleton.edu:81/
ehost/pdfviewer/pdfviewer?vid=9&hid=3&sid=965a771a-4662-47b08080-8da49b6f4146%40sessionmgr112
El Hajjar, S. T. (2011). An empirical study about the use and implementation of software in statistics at higher education institutions. International
Journal of Engineering & Technology, 11(6), 45–54. Retrieved from ProQuest database.
Gomez, R. (2010). Innovations in teaching undergraduate statistics courses
for biology students. Review of Higher Education and Self-Learning, 3(7),
8–13. Retrieved from ProQuest database.
Hamadu, D., Adeleke, I., & Ehie, I. (2011). Using information technology
in teaching of business statistics in Nigeria Business School. American
Journal of Business Education, 4(10), 85–92. Retrieved from ProQuest
database.
Harrington, D. (1999). Teaching Statistics: A comparison of traditional
classroom and programmed instruction/distance learning approaches.
Journal of Social Work Education, 35, 343–352. Retrieved from ProQuest
database.
Hawkes, J. S., & Marsh, W. H. (2005). Discovering statistics (2nd ed.).
Charleston, SC: Hawkes Learning Systems.
Hawkes Learning Systems. (2012). How the software works. Retrieved from
http://www.hawkeslearning.com/Instructors/HowTheSoftwareWorks.htm
Downloaded by [Universitas Maritim Raja Ali Haji] at 20:39 11 January 2016
BUSINESS STATISTICS: A COMPARISON
McLaren, C. H. (2004). A comparison of student persistence and performance in online and classroom business statistics experiences. Decision Sciences Journal of Innovative Education, 2, 1–10. http://dx.doi.org/
10.1111/j.0011-7315.2004.00015.x.
Mertler, C. A., & Vannatta, R. A. (2010). Advanced and multivariate statistical methods (4th ed.). Glendale, CA: Pyrczak.
Mills, J. D., & Raju, D. (2011). Teaching statistics online: A decade’s
review of the literature about what works. Journal of Statistics Education, 19, 1–27. Retrieved from http://www.amstat.org/publications/
jse/v19n2/mills.pdf
National Center for Education Statistics. (2011). Fast facts. Retrieved from
http://nces.ed.gov/fastfacts/display.asp?id=74
National Center for Higher Education Management Systems Information
Center for Higher Education Policymaking and Analysis. (2013). Educational levels of the population: Educational attainment by degree-level
and age-group (decennial census). Retrieved from http://www.highered
info.org/dbrowser/index.php?submeasure=201&year=2000&level=nat
ion&mode=data&state=0
Ott, R. L., & Longnecker, M. (2010). An introduction to statistical methods
and data analysis (6th ed.). Belmont, CA: Brooks/Cole–Cengage.
Radford, A. W. (2011, October). Learning at a distance: undergraduate enrollment in distance education courses and degree programs.
195
NCES 2012-154. Stats in Brief. Retrieved from http://nces.ed.gov/pubs
2012/2012154.pdf
Spinelli, M. A. (2001). The use of technology in teaching business statistics.
Journal of Education for Business, 77, 41–43. Retrieved from ProQuest
database.
Summers, J. J., Waigandt, A., & Whittaker, T. A. (2005). A comparison
of student achievement and satisfaction in an online versus a traditional
face-to-face statistics class. Innovative Higher Education, 29, 233–250.
http://dx.doi.org/10.1007/s10755-005-1938-x
Tallent-Runnels, M. K., Thomas, J. A., Lan, W. Y., Cooper, S., Ahern, T.
C., Shaw, S. M., & Liu, X. (2006). Teaching courses online: A review
of the research. Review of Educational Research, 76, 93–135. Retrieved
from ProQuest database.
Texas A&M University-Central Texas. (2012). Academic class
syllabi. Retrieved from http://www.tamuct.edu/departments/syllabi/
index.php
University of Phoenix. (2012). History. Retrieved from http://www.
phoenix.edu/about us/about university of phoenix/history.html
York, R. O. (2008). Comparing three modes of instruction in a graduate social work program. Journal of Social Work Education, 44, 157–172. Retrieved from ProQuest
database.