Academic success and the transfer of community college credits in the principles of economics.

Grimes, Paul W. ; Rezek, Jon P. ; Campbell, Randall C. 等

I. Introduction

In response to the escalating cost of college tuition at private institutions and public research universities, many of today's students choose to begin their higher educational experience at 2-year community colleges. During the decade between 1999 and 2009, enrollment in 2-year institutions rose by more than one and a half million students nationwide. Today, approximately 43 percent of all students enrolled in American institutions of higher learning attend 2-year schools (National Center for Educational Statistics 2010). Although no comprehensive census exists, many of these students will ultimately transfer to universities to pursue the traditional Bachelors degree. In fact, more than one half of all college students now earn credit hours at more than one institution prior to graduation (National Center for Educational Statistics 2010).

In many states, the major universities are eager to attract students who wish to transfer from local community colleges and regional institutions. It is common practice for major public universities to maintain articulation agreements with the community colleges and regional universities in their state and across their service region. These agreements serve as a guarantee that students who earn academic credits from their local institutions can transfer them into the major universities' degree programs. Today, many states mandate that articulation agreements must be reached between all in-state public institutions of higher learning.

For many students, financial constraints limit their ability to enroll in major universities and 2-year community colleges provide an affordable alternative. (1) However, casual empiricism suggests that some students choose to begin their higher education at community colleges not only in response to lower financial costs, but also due to perceptions of lower "effort costs." As a consequence some educators have openly questioned whether the rigor and quality of courses at community colleges are academically equivalent to those found on the campuses of major universities. Since the traditional introductory Principles of Economics courses are usually taught at the sophomore level, the question of course equivalency is particularly relevant to economic educators. Previous researchers in the field have addressed the issue, but the results are dated and may not reflect the current higher educational environment.

Following a national boom in the opening of new community colleges, which occurred in response to the baby boom generation's demand for higher education, a number of community college studies appeared in the economic education literature during the early 1970s. For example, Lewis, Wentworth and Orvis (1973) found that students at 2-year colleges performed significantly below their 4-year counterparts on the standardized Test of Understanding College Economics (TUCE). Additional research studies used 2-year institutions as the setting for analyzing the relative effectiveness of different teaching pedagogies (for example, Wentworth and Lewis (1973) and Ross (1977)).

More recently, Laband and Piette (1995) using data from the late 1980s and early 1990s, examined the academic performance of community college transfers with non-transfer (native) students at Florida State University. They found that the community college transfer students did indeed perform worse in upper division economics courses than the native students. In a series of articles, Hilmer (1997, 1999, 2002) addressed the more general issue of how previous college transfer experience affects the labor market earnings of university graduates. Hilmer's results support the human capital model of earnings and show that measures of institutional quality are reflected in graduates' wage profiles. Hilmer shows that student transfers between local/regional institutions and major universities reflect strategic behavior on the part of students seeking the higher returns to degrees granted by more prestigious universities. His empirical findings suggest that local colleges serve as accessible gateways for economically disadvantaged students to select into higher quality institutions.

To ultimately capture the economic benefits conferred by a major university degree, transfer students must receive instruction of a quality that prepares them to succeed at their destination institution. In this study we examine the role of 2-year community colleges in the higher education hierarchy with a particular emphasis on Principles of Economics. Specifically, this research addresses three primary questions. Are economics courses at 2-year community colleges academically equivalent to those taken at 4-year schools? Do academically-challenged students self-select into these colleges for economics courses? Do students who transfer general academic credit from 2-year community colleges achieve the same degree of academic success, measured in terms of overall grade performance, as those native students who earn equivalent credits in a university setting? (2)

II. Institutional Context and Data

This study uses Mississippi State University as a case example. MSU is a public Land Grant institution which enrolls a diverse student body with representatives from all 50 states and from more than 70 foreign nations. The largest university in the state, MSU has a total enrollment of approximately 20,400 full-time students. First-year students entering the university have an average ACT composite score of 23.6. MSU offers a full range of traditional academic programs from the Bachelors degree through the doctorate and is rated as a very high research activity institution by the Carnegie Foundation. In many ways, MSU reflects the institutional characteristics of a representative major public university in the United States. Over the past decade, MSU has experienced a significant increase in the number of students transferring in from community colleges, and to a lesser extent, from smaller regional state universities. As required by state regulations, MSU maintains articulation agreements with all of the state's fifteen 2-year and seven 4-year public institutions. Thus, our results may be generalized to a significant number of other universities with similar characteristics and environment.

Our sample consisted of all students who matriculated in the academic year 1997-1998 (including new freshman and transfer students) and who completed both courses of the traditional Principles of Economics sequence (Principles of Macroeconomics and Principles of Microeconomics) by academic year 2002-2003. (3) This length of time is often used for assessment and accreditation purposes to capture the six year graduation rate. A total of 892 students met the criteria for inclusion in the sample. Approximately 71 percent of the student subjects completed the Principles of Economics sequence at MSU with the remaining 29 percent transferring the credits from other institutions (23 percent from 2-year institutions and 6 percent from 4-year institutions). Native students who took their economics courses at MSU had an 82.5 percent graduation rate while only 68.5 percent of economics transfer students graduated in six years or fewer. (4)

Table 1 provides the basic descriptive statistics for the sample. The second column reports the means for the demographic characteristics, academic aptitude, and the final college of enrollment for students taking the Principles of Economics sequence at MSU. The third column provides the same information for students who took both of the Principles courses elsewhere and then transferred the credits to MSU. For all students in the sample with transfer credits on their official transcript, approximately 80 percent earned those credits at a 2-year community college. The final two columns delineate the transfer students by institution type.

For native and transfer students alike, Principles of Economics courses were disproportionately populated with white males compared to the institution as a whole. Only 39 percent of our sample was female and only 16 percent was African-American, while in the overall MSU student population 48.5 percent of students were female and 19.5 percent were African-American. Students transferring Principles of Economics credits from another institution were slightly younger and more likely to be white and from the state of Mississippi than all MSU economics students. Native economics students were more likely to major in a business field, but were less likely to major in accounting or an engineering field than the economics transfer students.

Academically, the average cumulative GPA, measured as the combination of the GPA earned at MSU and through transfer credits, was nearly identical across both subsets; however, the average ACT score was approximately 1.4 points higher among the native economics students than for the economics transfer students. These findings suggest that there may be a sample selection issue at work, with the decision to take economics at a community college or regional university being dependent on the student's academic abilities, as measured by their ACT score, and other demographic characteristics. Furthermore, the descriptive statistics indicate the grades earned in both macroeconomics and microeconomics were lower for students taking the course sequence at MSU relative to the grades earned by students who transferred the credit hours from other institutions. The combination of lower ACT scores but higher economics grades for transfer students suggests that grades may not be equivalent across institutions.

Differences also exist across the subsets in terms of the total transfer hours. The average MSU economics student transferred in about 18 semester credit hours in total. However, students who transferred credit for the Principles sequence averaged 71 hours credit at other institutions. Therefore, the economics transfer students were more likely to have completed an Associate's degree at a two-year community college. (5) Finally there does not appear to be a significant difference between when transfers and non-transfers took their Principles sequence. The students who transferred Principles credit likely took the courses in their first two years of college. This pattern is also observed for native Principles students, 77% of whom took these courses during their first two years at MSU.

III. Methodology and Results

We estimate an econometric model to determine the effect of transferring academic credits on ultimate academic success as measured by student grade point average. Specifically, the model takes the following functional form:

GPA = f(D, A, M, T) [1]

where GPA represents student academic performance as measured by cumulative grade point average (using the standard 4-point scale). (6) On the right-hand side, D is a vector of student demographic characteristics, A reflects student aptitude as measured by their ACT score and grades earned in the Principles of Economics course sequence, M represents the students' major field of study as reflected by the college of enrollment, (7) and T represents transfer credit. Note that this specification suffers from a classic self-sample selection problem since students choose whether or not to transfer Principles credits.

The empirical regression equation was specified as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] [2]

where,

MALE = 1 for male student, 0 for female student;

AFRICAN AMERICAN = 1 if student is black, 0 for other;

OUT-OF-STATE = 1 for out-of-state student, 0 for resident;

FOREIGN = 1 for international student, 0 for U.S. resident;

AGE = student's age in years

ACT = student's composite ACT score

ACCOUNTING = 1 if School of Accounting major, 0 otherwise;

ARTS & SCIENCES = 1 if College of Arts and Science major, 0 otherwise;

BUSINESS = 1 if College of Business major, 0 otherwise;

EDUCATION = 1 if College of Education major, 0 otherwise;

ENGINEERING = 1 if College of Engineering major, 0 otherwise;

TRANSFER HOURS = total number of transfer hours.

MACRO GRADE = grade in Principles of Macroeconomics (A = 4 ... F = 0)

MICRO GRADE = grade in Principles of Microeconomics (A = 4 ... F = 0)

Grade Point Average

The second panel of Table 2 reports the OLS estimation of equation [2]. The estimated equation yielded a significant F-statistic and an acceptable [R.sup.2] for cross-sectional data. Looking first at the results for the control variables, the coefficients for the race and gender variables were statistically significant at the .01 level. AFRICAN AMERICAN students had lower cumulative GPAs relative to non-black students and males had lower cumulative GPAs than females, ceteris paribus. The magnitudes of both of these coefficients were relatively large. The coefficient for AGE was also negative and significant at the .10 level, indicating that, holding all else constant, each decade of life reduced student performance by about 0.12 in cumulative GPA. This may reflect the increased opportunity costs facing many older students, or may indicate that time has eroded some of their academic skills. A student's country or state of origin did not significantly affect cumulative GPA.

As expected, student ACT scores were positively correlated with GPA and highly significant, although the magnitude was somewhat low. An increase of ten points in ACT score translated into an increase of only 0.29 in cumulative GPA. The low magnitude is possibly due to students with higher ACT scores taking more difficult courses; although we mitigate this effect somewhat by including dummies for the student's college, we cannot account for all differences in course difficulty. The results also show that grades in both the macroeconomics and microeconomics courses are strong predictors of eventual academic success in terms of cumulative GPA. A student receiving an A in Principles of Macroeconomics earned a 0.17 higher cumulative GPA than a student receiving a B. The effect was even larger for microeconomics, where an increase of one letter grade increased cumulative GPA by 0.21 points. These results are consistent with Grimes and Niss (1991) who found that economic understanding is strongly tied to ultimate academic performance.

The expected sign for the TRANSFER HOURS coefficient is uncertain a priori due to two possible effects associated with transferring credit. First, if the courses are less rigorous or are simply graded less rigorously, then students who transfer credits might be expected to earn higher cumulative GPA's since more of these 'easier' courses appear on their transcripts. Conversely, if students develop poor study habits from less rigorous coursework or simply do not learn the material necessary to succeed in upper division courses, their cumulative GPA could suffer once they transfer to a 4-year institution. Here we find the coefficient on the number of transfer hours to be insignificant, suggesting the effects may offset one another.

Table 2 also reveals that no significant difference existed between the performances of BUSINESS or ACCOUNTING students relative to those enrolled in OTHER colleges (the omitted category). Although neither of these college dummy variables were found to be significant, three college dummies were significant. The negative and significant ENGINEERING coefficient most likely reflects programs of study which involve advanced math and science courses typically viewed as being relatively rigorous. Conversely, GPAs in EDUCATION were higher than the baseline by about 0.17 points, which again may be attributed to the level of rigor in the required courses. Lastly, at least two explanations for the negative and significant coefficient for ARTS & SCIENCES majors are possible. First, the coursework in some Arts and Sciences fields may be more difficult relative to majors in other colleges, pushing grades lower. Second, some Arts and Sciences degree programs may serve as majors of last resort for students performing poorly in other colleges, also pushing GPAs down.

Course Equivalency between MSU and Transfer Institutions

The MACRO GRADE and MICRO GRADE coefficients, shown in panel 2 of Table 2, quantify the effect of a one letter grade improvement in each Principles course on eventual cumulative GPA. However, as specified these grade variables did not account for the location of where the courses were taken. To investigate the possibility that economics grades are inflated at transfer institutions, we incorporated a location effect by replacing the discrete course grade variables with a set of dummy variables representing where the course was taken (MSU, 2-year institution, or other 4-year institution) and the grade received. The magnitude of these dummy variables was used to compare the effects of Principles grades, across institutions, on overall academic achievement as measured by cumulative GPA.

Unfortunately, estimating a model which incorporates a full set of dummy variables for both macroeconomics and microeconomics by location created near perfect collinearity among the explanatory variables, therefore, we estimated two separate models, one in which the discrete MACRO GRADE variable was replaced with a series of macroeconomics grade/location dummies and one in which the discrete MICRO GRADE variable was replaced with a series of microeconomics grade/ location dummies. This modification of equation [2] yields the following two specifications:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] [3a]

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] [3b]

where,

MSU MACRO A = 1 for students receiving A's in macro at MSU, 0 otherwise

MSU MACRO B = 1 for students receiving B's in macro at MSU, 0 otherwise

MSU MACRO D = 1 for students receiving D's or F's in macro at MSU, 0 otherwise

T2 MACRO A = 1 for students receiving A's in macro at 2-year colleges, 0 otherwise

T2 MACRO B = 1 for students receiving B's in macro 2-year colleges, 0 otherwise

T2 MACRO C = 1 for students receiving C's in macro 2-year colleges, 0 otherwise

T2 MACRO D = 1 for students receiving D's or F's in macro 2-year colleges, 0 otherwise

T4 MACRO A = 1 for students receiving A's in macro at 4-year colleges, 0 otherwise

T4 MACRO B = 1 for students receiving B's in macro 4-year colleges, 0 otherwise

T4 MACRO C = 1 for students receiving C's in macro 4-year colleges, 0 otherwise

T4 MACRO D = 1 for students receiving D's or F's in macro 4-year colleges, 0 otherwise

and

MSU MICRO A = 1 for students receiving A's in micro at MSU, 0 otherwise

MSU MICRO B = 1 for students receiving B's in micro at MSU, 0 otherwise

MSU MICRO D = 1 for students receiving D's or F's in micro at MSU, 0 otherwise

T2 MICRO A = 1 for students receiving A's in micro at 2-year colleges, 0 otherwise

T2 MICRO B = 1 for students receiving B's in micro 2-year colleges, 0 otherwise

T2 MICRO C = 1 for students receiving C's in micro 2-year colleges, 0 otherwise

T2 MICRO D = 1 for students receiving D's or F's in micro 2-year colleges, 0 otherwise

T4 MICRO A = 1 for students receiving A's in micro at 4-year colleges, 0 otherwise

T4 MICRO B = 1 for students receiving B's in micro 4-year colleges, 0 otherwise

T4 MICRO C = 1 for students receiving C's in micro 4-year colleges, 0 otherwise

T4 MICRO D = 1 for students receiving D's or F's in micro 4-year colleges, 0 otherwise

Note that in both equations students earning a C at MSU are treated as the baseline group and are thus the omitted category.

The empirical results for [3a] and [3b] are presented in panels 3 and 4 of Table 2. As seen in the table, students receiving a B in macroeconomics at MSU are expected to earn a 0.16 higher GPA during their academic careers than baseline students, holding all other factors constant, while students receiving an A at MSU are expected to earn a 0.40 higher GPA than baseline students. Students receiving a D or F in macroeconomics at MSU are expected to earn a 0.16 lower GPA over the course of their academic careers than those earning C's. Each of these dummy variables is statistically significant at the .01 level.

The T2 coefficients quantify differences in cumulative GPA for 2-year economics transfer students relative to the baseline MSU C students. The results reveal that students transferring A's in macroeconomics from such schools are expected to earn a 0.16 higher GPA than the baseline students, holding other factors constant. The T2 MACRO B coefficient is near zero and insignificant, indicating that there is no significant difference in cumulative GPA between a student who earned a B in macroeconomics at a 2-year community college and a student who earned a C at MSU. The results further show the T2 MACRO C coefficient to be negative; however, it is not statistically significant at conventional levels, indicating that no conclusions can be made about the relationship between C's at MSU and 2-year colleges. Finally, students earning grades of D or F from 2-year schools are expected to earn a 0.35 lower cumulative GPA than baseline students.

The T4 coefficients quantify the difference in cumulative GPA for 4-year college transfer students relative to MSU C students. Macroeconomics students receiving A's from other 4-year institutions are expected to earn 0.36 higher cumulative GPAs than MSU's C students. However, the magnitude and standard error of the T4 MACRO B and T4 MACRO C coefficients indicate there is no statistical difference between baseline students and students transferring B's or C's from the other 4-year schools in the sample. Finally, the T4 MACRO D coefficient is negative 0.39, meaning students earning D's or F's in macroeconomics from four year schools perform significantly worse than baseline students.

Similar results hold for equation [3b] when microeconomics grade dummies replace the macroeconomics dummies, as shown in panel 4 of Table 2. Native economics students receiving a B in microeconomics are expected to earn a 0.17 higher cumulative GPA than baseline students, and A students are expected to earn a 0.43 higher cumulative GPA than baseline students. Students receiving a D or F are expected to earn a 0.24 lower cumulative GPA than MSU's C microeconomics students. Each of these dummy variables is statistically significant at the .01 level.

According to the estimated coefficients, student transferring A's in microeconomics from 2-year schools are expected to earn a 0.23 higher cumulative GPA relative to baseline students; and those transferring A's from 4-year schools are expected to earn a 0.39 higher GPA relative to baseline students, holding other factors constant. At conventional levels of significance, there is no statistical difference between B's earned at either 2- or 4-year institutions and C's earned in microeconomics at MSU. However, the results show that students transferring C's from 2-year institutions are expected to earn 0.17 lower cumulative GPAs than baseline students while C microeconomics students from other 4-year schools are expected to earn a 0.31 lower cumulative GPA. Finally, students earning D's or F's from all other institutions performed very poorly relative to baseline MSU students, with T2 MICRO D equal to -0.40 and T4 MICRO D equal to -0.36.

The empirical formulation of equations [3a] and [3b] allows us to create a measure of the equivalency between grades earned in the Principles of Economics courses at 2-year community colleges or other 4-year institutions and those earned at MSU. This measure is based on two simple premises. First, grades in Principles of Economics courses are good predictors of overall academic performance, and second, the predictive power of economics grades should be equal across institutions if instructors at those institutions teach similar curriculum and grade with relatively equal rigor. We define our grade equivalency measure as:

[E.sub.MT] = 1 - [([T.sub.i] - [M.sub.i-1])/([M.sub.i] - [M.sub.i-1])] [4]

where,

[E.sub.MT] is the grade equivalency between institution M and T

[T.sub.i] is the coefficient for the grade received at the transfer institution; i = A,B,C,D,

[M.sub.i] is the coefficient for the grade received at MSU; i = A,B,C,D.

Equation [4] is constructed such that when [T.sub.i] = [M.sub.i], then [E.sub.MT] = 0 and grades are considered equivalent across institutions. However, if [T.sub.i] = [M.sub.i-1] we obtain [E.sub.MT] = 1 and grades are inflated by a full letter grade at the transfer institution. If [E.sub.MT] > 1, then grades are inflated by more than one letter grade. For example, suppose we wish to calculate the equivalency of a 'B' grade received at a 2-year college and a 'B' received at MSU. From the results in Table 2, we calculate:

[E.sub.MT] = 1 - [([T.sub.B] - [M.sub.C])/([M.sub.B] - [M.sub.C])] = 1-[.00940/. 15966] = 0.9411.

The implied equivalencies between courses taken at 2-year schools or other 4-year institutions relative to MSU, as defined by equation [4], are given in Table 3. Results indicate that A grades in macroeconomics from 2-year colleges were inflated by nearly one full letter grade (0.997) but that there was only mild inflation in the A range at other 4-year institutions (0.178). Thus, an A in Principles of Macroeconomics obtained at a 2-year institution in our sample was almost identical to a B obtained at MSU, in terms of the overall impact on GPA and holding all else constant. For 4-year institutions in our sample, there was no significant difference between an A obtained at the transfer institution and an A obtained at MSU in macroeconomics. However, B's in macroeconomics at both 2- and 4-year institutions were inflated by about one letter grade as well (0.941 and 0.968, respectively), while C's were inflated by about 2/3 of a letter grade at 2-year schools and about 3/4 of a letter grade at other 4-year institutions.

The Principles of Microeconomics at 4-year institutions appears to be more equivalent to the course as taught at MSU, relative to its companion course in macroeconomics. No statistically significant difference was found for grades of A and B at other 4-year schools relative to MSU. However, C's in microeconomics were highly inflated relative to MSU. Table 3 shows that for 2-year community colleges, grade inflation ranged from 2/3 to 3/4 of a letter grade for microeconomics, somewhat less than macroeconomics but still considerable.

Sample Selection

Students may choose to take the Principles of Economics course sequence at a 2-year college for financial reasons or they may behave strategically, believing their chances of passing or receiving a higher grade at community colleges are greater. Given that their rewards may be greater, academically-challenged students may engage in such behavior disproportionately. To account for this possible sample selection phenomena, we re-estimated the GPA regression model using maximum likelihood techniques. As noted by Kennedy (2003, pp. 291-293), maximum likelihood estimation is an efficient alternative to Heckman's (1979) two-stage procedure for dealing with sample selection problems.

In our formulation, we developed a probit model to estimate the probability of a student selecting to take the Principles sequence at a 2-year community college in which student aptitude, as proxied by ACT scores, and observable demographic characteristics served as explanatory variables. (8) Specifically, we used the LIMDEP statistical package to simultaneously estimate equation [3a] (or [3b]) and the following selection equation:

ECON = [[alpha].sub.0] + [[alpha].sub.1]MALE + [[alpha].sub.2] AFRICAN AMERICAN + [[alpha].sub.3]OUT-OF-STATE + [[alpha].sub.4]AGE + [[alpha].sub.5]ACT + [epsilon] [7]

where,

ECON = 1 if the courses were taken at a 2-year institution, 0 if the courses were taken at MSU.

The resulting probit estimates for choosing to take the Principles sequence at a 2-year community college are shown in Table 4. These maximum likelihood estimates revealed that students opting for economics at such colleges were disproportionately white state residents, younger when they took their economics classes, and most importantly, scored significantly lower on the ACT than their counterparts taking economics at MSU or other 4-year institutions. This last result suggests that less academically-able students self-selected into 2-year community colleges for the Principles of Economics course sequence and then transferred to MSU.

Accounting for sample selection resulted in only slightly different results for the GPA equations, as shown in Table 5. The only significant differences in the GPA equations after adjusting for sample selection bias were in the 2-year transfer dummy variables for macroeconomics grades. The OLS version of the model indicated no significant difference between a C in macroeconomics at MSU and a C at 2-year community colleges. The adjusted results show that students receiving a C at a 2-year school are expected to earn a 0.14 lower cumulative GPA than baseline students. This difference is significant at the .10 level. Similarly, according to the adjusted results, there is no longer a significant difference between a C in macroeconomics at MSU and an A in macroeconomics at a 2-year school in terms of the resulting cumulative GPA.

Finally, TRANSFER HOURS appears positive and significant in both the adjusted and unadjusted models. However, in all cases, the magnitude appears to be quite small (approximately 0.0013). Thus, if a student took 60 hours at a 2-year college, the overall improvement in cumulative GPA was only about 0.08. While this appears to be quite low, recall that two effects are intertwined in this coefficient. "Easier" courses would tend to make this coefficient more positive, but if these courses leave students less prepared for higher level courses, then more hours at a 2-year college would hinder a student's overall GPA, making this coefficient more negative. Since this paper provides evidence that Principles of Economics grades may be inflated at 2-year institutions, the low value of this coefficient suggests that the latter 'preparedness' effect may be counteracting the former 'grade inflation' effect.

Table 6 provides the grade equivalency measures based on the sample selection corrected results. The pattern of statistical significance is identical to those obtained when the self-selection issue was not taken into account. However, in all cases, the relative magnitudes of the corrected equivalency measures are greater than the uncorrected measures shown in Table 3. This provides additional evidence for the existence of strategic behavior by students with relatively weaker academic aptitudes--transferring credit from a 2-year community college may significantly increase cumulative GPA, and by extension ultimately increase the likelihood of graduation from a major university.

IV. Summary and Conclusions

Our analysis of academic success for students who transfer credit in the Principles of Economics from local community colleges and regional universities to a major research university revealed several interesting and important results. First, when we just control for the grade earned in both Principles classes, regardless of the institution at which they were taken, we find that both courses are significant predictors of overall academic success as captured by cumulative GPA. Although statistically insignificant, students who transferred more credits earned lower GPA's, however, when we included dummy variables indicating both the grade received and where the Principles courses were taken, we found that TRANSFER HOURS had a positive and significant effect on cumulative GPA. This suggests that students of equal ability received higher grades if they took the courses at a 2-year college. Thus, we developed a grade equivalency measure based on our regression results, and found that Principles of Economics grades at 2-year community colleges were inflated by nearly a full letter grade for both the macroeconomics and microeconomics course.

Since students self-select into 2-year institutions, we estimated the model again to account for sample selection bias. We found that students who chose to transfer from local community colleges had significantly lower ACT scores and, therefore, were less prepared academically for future studies. Again, the results suggested that grades in the principles courses were between 3/4 and 5/6 of a letter grade higher in 2-year institutions compared to MSU. Our model indicated that the transfer of credits from 2-year institutions is a rational economic choice in that it raises cumulative GPAs, and by extension, enhances the probability of eventual graduation for this set of students.

Our results indicated that in addition to serving as a gateway for economically disadvantaged students, local community colleges serve as a gateway for less academically qualified students. These findings are important given the growing number of students who begin their higher education experience at local community colleges and then transfer to a major public university. Economic educators at institutions receiving transfer students should be aware of the self-selection process revealed by our findings and be cognizant that such students may need additional attention in order to achieve academic success. Furthermore, since there appears to be a significant lack of equivalency between Principles of Economics course grades between types of institutions, articulation agreements may need to be reevaluated to determine if credit should be accepted for students who earned relatively low grades in Principles courses from 2-year community colleges. Alternatively, university economics departments that experience a significant number of 2-year transfers could consider the implementation of placement exams similar to those commonly found in departments of mathematics and foreign languages.

The issues surrounding course and grade equivalency highlighted by our findings also suggest that it would be helpful for university economists to open a dialog with their colleagues at local 2-year community colleges. If common standards could be agreed upon and adopted by sending and receiving institutions, students may be better prepared for future academic studies and grades more accurately reflect student proficiencies. Clearly, additional research is needed to determine exactly what type of practices and policies will enhance the ultimate success rate for community college transfer students.

References

Grimes, Paul W., and Niss, James F. 1991. Economic understanding and student success in a business curriculum. Journal of Education for Business, 66(5), 309-313.

Heckman, James. 1979. Sample bias as a specification error. Econometrica, 47(1), 153-162.

Hilmer, Michael J. 1997. Does community college attendance provide a strategic path to higher quality education? Economics of Education Review, 16(1), 59-68.

Hilmer, Michael J. 1999. Does the return to university quality differ for transfer students and direct attendees? Economics of Education Review, 19(1), 47-61.

Hilmer, Michael J. 2002. Human capital attainment, university quality, and entry-level wages for college transfer students. The Southern Economic Journal, 69(2), 457-469.

Kennedy, Peter. 2003. A Guide to Econometrics, 5th edition. Cambridge, MA: MIT Press.

Laband, David, and Piette, Michael. 1995. Does who teaches principles of economics matter? American Economic Review, 85(2), 335-338.

Lewis, Darrell R., Wentworth, Donald R., and Orvis, Charles C. 1973. Economics in junior colleges: Terminal or transfer? Journal of Economic Education, 4(2), 100-110.

National Center for Educational Statistics. 2010. Digest of Educational Statistics, 2010. Washington, DC: United States Department of Education.

Ross, Thomas. 1977. Macromod: Computer simulation for introductory economics. Social Science Journal, 1(2), 26-29.

Wentworth, Donald R., and Lewis, Darrell R. 1973. An evaluation of a simulation game for teaching introductory economics in junior colleges. Journal of Experimental Education, 42(2), 87-96.

Notes

(1.) For instance, in 2011-12 the full-time tuition rate for state residents at Mississippi State University (MSU) was $2,902 but only $2,000 for the state's 2-year schools.

(2.) A question of related interest to economic educators whether students who transfer Principles of Economics credits from 2-year community colleges achieve the same degree of success in subsequent economics courses as those students who earn equivalent credit at the home institution. This question was previously investigated by Laband and Piette (1995) but poses two problems here. First, our dataset does not delineate GPAs by major (for many majors, not all course requirements carry the discipline's course prefix) and second, the number of economics majors in the sample was not large enough to make any statistically significant conclusions. (Recall that the sample was defined as those who students who matriculated during one academic year.)

(3.) As at many institutions, the Principles of Economics course sequence is often taken by MSU students as part of the university's core curriculum requirements. In any given semester, approximately 1,000 students enroll in the two courses combined. Given the institutional arrangements, many students in majors outside of the business school take only one of the two courses. Our sample, which includes only those students who took both courses, was defined to ensure that we captured students with a strong academic interest in successful completion of the course sequence.

(4.) A total of 124 credit hours with a cumulative GPA of 2.0 or better is required for graduation at MSU, although some colleges, including the College of Business, require a higher standard. Among the 192 students who did not graduate, the data show that the majority were not close to finishing. Of the non-graduates, 49.5% had a cumulative GPA less than 2.25 while 51% had an MSU GPA below 2.0. Nearly one-third of these students (31.8%) had fewer than 90 credit hours while over half (53.6%) had fewer than 110 credit hours. Finally, only 41 of the 192 non-graduates were enrolled during the 2002-2003 academic year. Thus, the majority of non-graduates had stopped making progress toward a degree at the time the study was concluded.

(5.) In Mississippi, the State College Board regulations require a minimum of 124 semester hours of credit to be earned for all Bachelors degrees. A minimum of 60 semester hours are required for all Associates degrees granted by the state's community colleges.

(6.) We chose cumulative GPA rather than MSU GPA to measure academic performance for two primary reasons: 1) According to the official MSU Bulletin, graduates must "make an overall C (2.0) average on all hours scheduled and rescheduled at all institutions attended, including Mississippi State University." Therefore, to receive an undergraduate degree, a student must earn a minimum 2.0 for all transfer and all MSU hours. Thus, whether or not a student ultimately graduates is dependent upon the cumulative GPA which accounts for both of these separate measures. A minimum 2.0 cumulative GPA is a necessary, though not sufficient, condition of graduation. (MSU's minimum graduation policy is common to a vast majority of its peer institutions in the Southeast and Midwest.) 2) The cumulative GPA includes a greater number of hours and thereby offers a more meaningful measure of overall academic performance. There are several transfer students in the sample who took only a few (<30) hours credit at MSU. In our opinion, a GPA measured on such a small number of hours is not very meaningful. However, in addition to the reported results using cumulative GPA, we ran each model using MSU GPA for the full sample and using MSU GPA but deleting those observations with 30 or fewer MSU credit hours. The results for these models are very similar to the reported results and are available upon request.

(7.) The major academic administrative divisions of MSU include the College Agriculture and Life Sciences, the College of Architecture, Art, and Design, the College of Arts and Sciences, The College of Business, The College of Education, The Bagley College of Engineering, and the College of Forestry. The omitted OTHER reference category is the agglomeration of the agriculture, architecture, and forestry colleges which have relatively low economics enrollments.

(8.) We also ran the model using all transfers as the dependent variable and obtained similar results. When we run the model using Heckman's (1979) two stage procedure, the coefficient for LAMBDA is negative and significant.

by Paul W. Grimes *, Jon P. Rezek **, and Randall C. Campbell ***

* Dean and Professor of Economics, Kelce College of Business, Pittsburg State University, 101 Kelce Center, Pittsburg, KS 66762, and Emeritus Professor of Economics, College of Business, Mississippi State University, Mississippi State, MS 39762, E-mail: paul.grimes@pittstate.edu

** Associate Professor of Economics, College of Business, Mississippi State University, Box 9580, Mississippi State, MS 39762, E-mail: jrezek@cobilan.msstate.edu

*** Associate Professor of Economics, College of Business, Mississippi State University, Box 9580, Mississippi State, MS 39762, E-mail: rcampbell@cobilan.msstate.edu

The authors sincerely thank the late Peter Kennedy and anonymous referees for constructive comments and suggestions. Thanks are also extended to John O'Bannon, formerly of Mississippi State University's Information Technology Services, for compilation of the student data records needed to complete this research. Marybeth Grimes provided expert editorial assistance.

TABLE 1.
Descriptive Statistics: Sample Means and Standard Deviations by
Transfer Group

 Native MSU All Transfers

Dependent Variables
 GPA (4-point scale) 2.92 2.92
 (0.57) (0.62)
 GRADUATION 82.52 68.48
Demographics--D
 MALE (%) 61.26 61.09
 AFRICAN 17.95 11.28
 AMERICAN (%)
 OUT-OF-STATE (%) 19.21 7.39
 INTERNATIONAL (%) 2.52 0.39
 AGE (in years) 20.56 19.86
 (2.06) (1.73)
Academic Aptitude--A
 ACT Score 22.35 20.94
 (Comprehensive) (4.36) (4.31)
 Avg. Macroeconomics 2.75 2.88
 Grade
 Avg. Microeconomics 2.64 3.01
 Grade
Major College--M
 Accounting * (%) 7.24 10.12
 Arts and Sciences (%) 7.09 8.95
 Business (%) 57.80 48.64
 Education (%) 12.28 12.06
 Engineering (%) 8.66 10.12
 Other (%) 6.93 10.12
Transfer Credit--T
 Transfer Hours 17.78 70.99
 (24.05) (22.59)
Total Students 635 257

 2-Year School 4-Year School
 Transfers Transfers

Dependent Variables
 GPA (4-point scale) 2.89 3.06
 (0.60) (0.67)
 GRADUATION 67.98 70.37
Demographics--D
 MALE (%) 64.04 50.00
 AFRICAN 9.85 16.67
 AMERICAN (%)
 OUT-OF-STATE (%) 6.40 11.11
 INTERNATIONAL (%) 0.49 0.00
 AGE (in years) 19.87 19.83
 (1.74) (1.71)
Academic Aptitude--A
 ACT Score 20.56 22.37
 (Comprehensive) (4.14) (4.84)
 Avg. Macroeconomics 2.88 2.87
 Grade
 Avg. Microeconomics 3.00 3.06
 Grade
Major College--M
 Accounting * (%) 7.88 18.52
 Arts and Sciences (%) 8.87 9.26
 Business (%) 51.23 38.89
 Education (%) 10.34 18.52
 Engineering (%) 10.84 7.41
 Other (%) 10.84 7.41
Transfer Credit--T
 Transfer Hours 69.68 75.9
 (21.42) (26.16)
Total Students 203 54

Standard deviations are in parenthesis.

* The School of Accountancy is a unit of the College of Business at
MSU.

TABLE 2.
OLS Results--GPA Equations

 Equation 2 Equation 3a

 Variable [beta] S.E. [beta] S.E.

 CONSTANT 1.43733 0.16807 1.92979 0.17515
Demographics--D
 MALE -0.22835 0.0267 -0.22745 0.02677
 AFRICAN AMERICAN -0.21548 0.03812 -0.21943 0.03827
 OUT-OF-STATE -0.02059 0.03659 -0.01669 0.03674
 FOREIGN 0.15645 0.09774 0.12847 0.09800
 AGE -0.00465 0.00641 -0.01213 0.00671
Academic Aptitude--A
 ACT 0.03098 0.00372 0.02923 0.00379
 MACRO GRADE 0.17498 0.01669
 MSU MACRO A 0.39970 0.04603
 MSU MACRO B 0.15966 0.03814
 MSU MACRO D -0.15514 0.05761
 T2 MACRO A 0.16032 0.06741
 T2 MACRO B 0.00940 0.06067
 T2 MACRO C -0.09935 0.06474
 T2 MACRO D -0.34687 0.10959
 T4 MACRO A 0.35686 0.09310
 T4 MACRO B 0.00515 0.11285
 T4 MACRO C -0.11569 0.11822
 T4 MACRO D -0.39356 0.15256
 MICRO GRADE 0.20597 0.01559 0.21648 0.01570
 MSU MICRO A
 MSU MICRO B
 MSU MICRO D
 T2 MICRO A
 T2 MICRO B
 T2 MICRO C
 T2 MICRO D
 T4 MICRO A
 T4 MICRO B
 T4 MICRO C
 T4 MICRO D
Major College--M
 ACCOUNTING 0.05163 0.06477 0.05067 0.06498
 ARTS & SCIENCES -0.13904 0.06418 -0.12556 0.06444
 BUSINESS 0.03372 0.04848 0.04193 0.04854
 EDUCATION 0.18815 0.05799 0.18709 0.05850
 ENGINEERING -0.11681 0.06414 -0.11583 0.06433
Transfer Credit--T
 TRANSFER HOURS -0.00009 0.00042 0.00131 0.00060
Adjusted R-squared 0.58576 0.58833
F-statistic 90.99570 54.05623

 Equation 3b

 Variable [beta] S.E.

 CONSTANT 2.01986 0.17610
Demographics--D
 MALE -0.22842 0.02673
 AFRICAN AMERICAN -0.21448 0.03829
 OUT-OF-STATE -0.01253 0.03679
 FOREIGN 0.13489 0.09796
 AGE -0.01215 0.00674
Academic Aptitude--A
 ACT 0.02923 0.00378
 MACRO GRADE 0.18235 0.01676
 MSU MACRO A
 MSU MACRO B
 MSU MACRO D
 T2 MACRO A
 T2 MACRO B
 T2 MACRO C
 T2 MACRO D
 T4 MACRO A
 T4 MACRO B
 T4 MACRO C
 T4 MACRO D
 MICRO GRADE
 MSU MICRO A 0.43013 0.04636
 MSU MICRO B 0.16648 0.03841
 MSU MICRO D -0.23534 0.04919
 T2 MICRO A 0.23201 0.06255
 T2 MICRO B 0.06148 0.06159
 T2 MICRO C -0.17440 0.06880
 T2 MICRO D -0.40280 0.10722
 T4 MICRO A 0.39422 0.08835
 T4 MICRO B 0.12599 0.11024
 T4 MICRO C -0.31288 0.13436
 T4 MICRO D -0.36318 0.16219
Major College--M
 ACCOUNTING 0.05240 0.06512
 ARTS & SCIENCES -0.13032 0.06430
 BUSINESS 0.04244 0.04867
 EDUCATION 0.18125 0.05836
 ENGINEERING -0.11130 0.06439
Transfer Credit--T
 TRANSFER HOURS 0.00125 0.00060
Adjusted R-squared 0.58768
F-statistic 53.91465

TABLE 3.
Measures of Course Equivalency

 Macroeconomics

Grade 2-year 4-year

A 0.9973 *** 0.1785
B 0.9411 *** 0.9677 ***
C 0.6404 *** 0.7457 ***

 Macroeconomics

Grade 2-year 4-year

A 0.751 *** 0.136
B 0.631 *** 0.243
C 0.741 *** 1.329 ***

*** Grade equivalency rejected at the .01 level

TABLE 4.
Probit Results for Selection Equation (Dependent
Variable: ECON)

 Variable Probit Standard
 Coefficient Error

 CONSTANT 5.25509 0.81295
Demographics--D
 MALE 0.04899 0.10329
 AFRICAN AMERICAN -0.77984 0.16091
 OUTSTATE -0.67068 0.15810
 AGE -0.19517 0.03411
Academic Aptitude--A
 ACT -0.08804 0.01317

Log likeligood function: -424.8268

TABLE 5.
MLE Results--GPA Equations

 Variable Equation 3a Equation 3b

 [beta] S.E. [beta] S.E.

 CONSTANT 1.88126 0.15923 1.96897 0.15861
Demographics--D
 MALE -0.22806 0.03030 -0.22886 0.03037
 AFRICAN AMERICAN -0.20977 0.04192 -0.20428 0.04179
 OUT-OF-STATE -0.00923 0.03810 -0.00472 0.03840
 FOREIGN 0.13122 0.12409 0.13780 0.12193
 AGE -0.01163 0.00530 -0.01161 0.00524
Academic Aptitude--A
 ACT 0.03002 0.00391 0.03005 0.00384
 MACRO GRADE 0.18256 0.01695
 MSU MACRO A 0.40364 0.04746
 MSU MACRO B 0.16185 0.04080
 MSU MACRO D -0.15503 0.05689
 T2 MACRO A 0.11309 0.08456
 T2 MACRO B -0.02770 0.07517
 T2 MACRO C -0.13816 0.07419
 T2 MACRO D -0.37837 0.10913
 T4 MACRO A 0.35460 0.12302
 T4 MACRO B 0.00224 0.10296
 T4 MACRO C -0.11700 0.10028
 T4 MACRO D -0.40030 0.13498
 MICRO GRADE 0.21608 0.01543
 MSU MICRO A 0.43244 0.04629
 MSU MICRO B 0.16642 0.04119
 MSU MICRO D -0.23533 0.04751
 T2 MICRO A 0.18354 0.07507
 T2 MICRO B 0.01984 0.07604
 T2 MICRO C -0.21582 0.07453
 T2 MICRO D -0.42985 0.12142
 T4 MICRO A 0.39225 0.10708
 T4 MICRO B 0.12338 0.10912
 T4 MICRO C -0.32189 0.12328
 T4 MICRO D -0.37113 0.13247
Major College--M
 ACCOUNTING 0.04954 0.06992 0.05059 0.07032
 ARTS & SCIENCES -0.12866 0.05457 -0.13350 0.05455
 BUSINESS 0.03939 0.04284 0.03981 0.04265
 EDUCATION 0.18668 0.05400 0.18066 0.05349
 ENGINEERING -0.11665 0.06296 -0.11201 0.06327
Transfer Credit--T
 TRANSFER HOURS 0.00133 0.00061 0.00127 0.00061

TABLE 6.
Measures of Course Equivalency (Controlling for
Sample Selection)

 Macroeconomics

Grade 2-year 4-year

A 1.2017 *** 0.2028
B 1.1711 *** 0.9862 ***
C 0.8912 *** 0.7547 ***

 Macroeconomics

Grade 2-year 4-year

A 0.9357 *** 0.1510
B 0.8808 *** 0.2586
C 0.9171 *** 1.5771 ***

*** Grade equivalency rejected at the .01 level