wo areas of investment in higher education that are at the discretion of states are appropriations to higher education and state student aid. For the purposes of this section, these two areas are thought of as the "levers" of state investment. They are also two areas of input around which much state policy discussion revolves. The intent of this section is to investigate patterns and trends in Measuring Up 2000 state grades according to the varying degrees of higher education investment. Numerous studies offer useful commentary on the relationship between student aid and participation or appropriation levels and enrollment, but this section analyzes the relationship of appropriations and student aid to Measuring Up 2000 grade data only.
To better capture comparative data, state investments were characterized in two steps. First, rather than using total appropriations to higher education and total student aid by state, the analysis uses appropriations per student and student aid per student. This eliminates wide variations in state investment data due to differences in magnitude of actual dollar amounts. It also provides a better basis for relative comparison, since the dollar amount is divided by a common demographic factor. Second, states were not grouped together by region, but by quintile according to the degree of investment per student. Thus, the ten states with the highest level of student aid per student were ranked in the first quintile, the next ten states with the second highest level of student aid per student were ranked in the second quintile, and so on. The same procedure was followed for appropriations per student. For reference purposes, state groupings by quintile for student aid per student and appropriations per student are shown in Appendix C.
In this section, analysis is included only where state input in relation to a specific report card category produced a statistically significant relationship. The complete correlations between appropriations per student and all report card categories, and between student aid per student and all report card categories, are shown in Appendix D.
The first graphics in this section are between the state inputs and the completion category only. This is because completion was the only category that yielded significance in relation to both state inputs. Important observations about each graphic are highlighted. However, these suggestions and explanations only reflect the author's interpretation of the data. The more important purpose of the suggestions and explanations is to encourage discussion about the usefulness and merit of using such a graphic as a supplement to the report card. Would such information provide insight for states trying to improve their grades?
Data Sources for this Section
Each section of this exploratory report was intended to create discussion to help advance ideas that might be useful for a supplement to the report card. Thus, the overall purpose of this section is to analyze the value of comparing state appropriations and state student aid data to the results in Measuring Up 2000. Nationally available data was used as a starting point for this exploration, and hence broad state-level indicators often contain assumptions that must be further scrutinized. State appropriations and state student aid data are for 1999-2000 and 1998-1999, respectively. The data source is the Chronicle of Higher Education Almanac, 2000-2001 issue. According to the Chronicle, state appropriations figures are from Illinois State University and include state tax funds appropriated for colleges and universities, but do not include capital outlays and money from other sources, such as appropriations from local governments. This exclusion of local money must be noted, since the broad state-level analysis in this section combines state appropriations with report card grades. Participation, affordability and completion all contain subcategories that include community colleges, yet the state appropriations data is missing an important funding component for community colleges.
This analysis, then, speaks only to state investments and the results (according to report card performance) a state obtains because of those investments. Local investments are unquestionably influential in performance, however, particularly in states with large community college systems. The effect of this local investment is not captured in this analysis, but with further work the analysis could be refined to combine state and local investment.
COMPLETION GRADES BY STATE STUDENT AID PER STUDENT
Figure 6 displays, graphically, all state completion grades in relation to quintile by level of student aid per student, with the ten states investing the highest amount of state aid per student falling in the first quintile, and so forth. (See Appendix C for a listing of states by quintile category.) Figure 6 is not intended to communicate specific state grades but rather trends, variations and comparisons among and between quintile groups. The graphic also has the advantage of displaying the grade dispersion by quintile. The level of a state's investment in student aid per student showed a significant positive correlation with completion grades (Appendix D), and the trend line shown in Figure 6 is consistent with the correlation results. State investment in student aid per student did not yield any other significant correlation with the other four state higher education graded categories.
The trend line shown in Figure 6 displays the general pattern of student aid investment to report card completion. In general, states that invest heavily in student aid performed better on the completion category than states with lower investments. Northeastern states were disproportionately represented in the first quintile in Figure 6, and Western states were disproportionately represented in the fifth quintile. The lowest grade for any state appearing in the first quintile in Figure 6 was a C+, Illinois (index score 79), but New Hampshire, a Northeastern state, received an A (101) despite being in the fifth quintile in state aid per student. Utah received the lowest grade (D+, or 68) of all states appearing in the fifth quintile in Figure 6.
COMPLETION GRADES BY STATE APPROPRIATIONS PER STUDENT
Figure 7 displays the general pattern of higher education appropriations per student to report card completion for all 50 states. This correlation was significant. The relationship between appropriation level and completion grade would seem counterintuitive. In general, states that appropriate greater amounts per student fare worse on the completion category than states with lower investments. Interestingly, the report card categories affordability and completion are also negatively correlated.
Regional differences in Figure 7 (and Appendix C) are prominent, however. Five of the states that appropriate the lowest amount per student to higher education (in the fifth quintile) in Figure 7 are Northeastern states, and all of those states received an A grade in completion (90-101). Five Southern states appeared in the first quintile, and those states had varying grades. Three Western states were in the first quintile, with completion grades of C, D-, and F. These observations may also hint at underlying state philosophies regarding how higher education should be funded. More analysis and qualitative context on an individual state basis would more solidly ground such observations to help policymakers better understand their state's performance.
COMBINING STATE STUDENT AID AND APPROPRIATIONS
Student aid and appropriations data were also analyzed together. Table 6 shows those states that appeared in the first quintile in both student aid and higher education appropriations, and those states that appeared in the fifth quintile in both student aid and higher education appropriations. States that ranked in the first quintile in one grouping but the fifth in the other are also displayed. The corresponding completion grades for each state also are given in Table 6.
The completion results in Table 6 may be influenced by regional differences or individual state outliers, but some general interpretations still surface from Figures 6 and 7, Table 5, and Appendix C. Two Southern states appeared in the first quintile in both student aid and higher education appropriations per student. These states maintained very respectable completion grades by investing generously both in students and directly in higher education. With the exception of New Hampshire, a Northeastern state, states that fell in the fifth quintile in aid and appropriations had mediocre results. Three of the five states that fell in this category were Western states, and the highest grade obtained by one of these Western states was a C (73).
The two states that were in the first quintile in aid but in the fifth quintile for appropriations received A's, and both were Northeastern states. Two Southern and one Western state heavily invested in appropriations but were in the fifth quintile for student aid. On the whole, these states maintained slightly better than average results.
Some general observations can be drawn from this analysis. First, those states that appeared in the first quintile in student aid investment did well, regardless of appropriations quintile placement. Only one state in the first quintile for student aid (Illinois) received a C+ for completion; the remainder received a B- or higher. The results in Table 6 regarding those states that invest heavily in student aid coincide with the correlation analysis. This suggests that the level of student aid per student is very important in directly improving completion rates.
From Table 6 and Appendix C, it appears that low levels of appropriations do not doom a state to a low completion grade, nor do high levels of appropriations guarantee a high grade. That states in the top quintile listed in Table 6 in appropriations per student did obtain slightly better than average completion grades somewhat contradicts the correlation results, but it does provide anecdotal evidence that strong investment helps completion. These results may be because select states from the first quintile group are represented in Table 6, however. The entire first quintile group for appropriations (in Appendix C) does show wide variation in completion grades, with many states scoring in the D (60s) range.
Many factors influence how much a state invests in appropriations and/or student aid. For example, the obvious importance of private institutions in the Northeastern region of the United States is part of the context that must be taken into consideration when discussing different policy strategies that might influence completion. To that end, however, a general conclusion for the nation as a whole might be that high levels of student aid provide more leverage to increase one's completion rate than do high levels of higher education appropriations per student.
To ensure that these observations are within reason, a correlation analysis was also performed between the two state inputs (aid and appropriations) and all regions excluding the Northeast (Appendix D). (Another correlation analysis was conducted between the two state inputs and one region, the West [also in Appendix D], because this is the region of the sample state, New Mexico.)
These additional analyses were performed because if any of the results showed significance, and the nature of that significance was different from the correlation analysis that included all regions, then any interpretations should be accompanied with extreme caution. Such a case would be an indicator that regional data were disproportionately influencing national data analysis.
It is still advisable to take caution in making broad interpretations of the data, but the additional correlation analyses did not yield any results that contradicted the analyses that included all regions. The correlation analysis for the Western region produced no significance, perhaps because of the reduced number of observations, which in turn would have had to yield a much stronger connection between the report card grades and state inputs to yield a significant result.
INCLUDING AFFORDABILITY AND PARTICIPATION
The relationships and differences between affordability, student aid, and appropriations are of central importance. First, high investment in student aid per student does not guarantee affordability. According to the data in this analysis, there is no significant correlation between student aid and affordability. Student aid was positively correlated to completion. Appropriations per student has a very different relationship with the report card grades. Appropriations per student is negatively correlated with completion and participation, but positively correlated with affordability. Indeed, those states that were grouped in the first quintile for appropriations, as shown in Appendix C, had very mixed results in completion, participation, and affordability.
States in the first quintile for student aid showed mixed results on participation and affordability, though most did well on completion. Of the four states in the top quintile for student aid shown in Table 6, only North Carolina received a strong grade for affordability. One is left with the impression that heavy investment in state student aid helps with completion but doesn't make college more affordable nor opportunities any more available for those who need it the most.
The next question I addressed directly placed participation rates at the center of the discussion. Do low grades in affordability hurt participation? And what is the effect of higher education appropriations and student aid on participation? Most importantly, how should a state balance appropriations and student aid to maintain participation, affordability, and completion?
This analysis gives no indication that either affordability or student aid impact participation. Direct appropriations to higher education per student actually has an inverse relationship with participation, from the report card data. Southern states were disproportionately represented (6 out of 10 states) in the first quintile of higher education appropriations per student, but only one Southern state in this quintile received a C+ on participation while the rest received D's and lower. Given that regional differences and state outliers do exist and may influence national data trends, the larger question is how states properly balance the two inputs of appropriations and student aid to maximize participation, affordability and completion.
CONCLUSION: MAXIMIZING PARTICIPATION, AFFORDABILITY AND COMPLETION
For the purposes of this section, benefits and preparation categories were not included, as neither showed significance to either of the two state inputs. Participation, affordability and completion were correlated to at least one of the state inputs. This section is concerned with analysis to investigate whether or not there is an ideal balance between aid per student and appropriations per student that would maximize a state's report card grades in participation, affordability, and completion. Such an investigation is concerned with addressing this question for the nation as a whole. Regional and state differences will be outlined to the extent that major issues, outliers, or caveats need to be highlighted. These cases will surface most prominently in the individual state analysis, the final section of this report, for which New Mexico serves as an example.
To approach the question of state inputs and grade maximization, I looked at the three aforementioned report card categories simultaneously and compared them to the two state inputs. Appendix E shows each state's three-grade average for participation, affordability and completion. States were then sorted by the highest average for the three report card categories (Appendix F), by quintile. Thus, the ten states with the highest three-grade average were ranked in the first quintile, the next ten states with the second highest three-grade average were ranked in the second quintile, and so on.
An artificial ratio of student aid per student divided by appropriations per student was then generated to compare the "mix" of aid and appropriation strategy by averages, to see if any pattern emerged. The results of the ratio are also shown in the table in Appendix F, along with the states sorted by highest three-grade average. The ratio is multiplied by a factor of 100 just to scale each of the scores so that they are not too small and thus difficult to read and compare. The national average was 4.72. A ratio value above 4.72 indicates that the state devotes more to student aid relative to appropriations per student than the national state average. A ratio value below 4.72 indicates that the state devotes less to student aid relative to appropriations per student than the national state average. Table 7 summarizes the analysis in Appendices E and F.
For the most part, quintiles that favor aid in the aid-to-appropriations ratio seemed to be the most successful at reaching or exceeding the three-grade national average. Five Northcentral states surfaced in the top quintile of total average performance, while Southern states were disproportionately represented in the bottom two quintiles. The second quintile contained four Northeastern states but was still below the national ratio average of 4.72. This quintile's average ratio value was brought down significantly by two Western states with very low ratio values (Wyoming and Utah). The fourth and fifth quintiles had averages that were well below the national average.
From the perspective of a quintile analysis, there is no discernable pattern of what mix of aid to appropriations yields a strong three-grade average. The ratio values for the first and third quintiles strongly favor an emphasis on aid, relative to the national average. The third quintile group was close to the three-grade national average, and the first quintile group easily exceeded it. The second quintile group exceeded the three-grade national average but favored appropriations over aid. What Table 7 does seem to indicate is that when the ratio value falls significantly below the national ratio average, total average performance suffers. The bottom two quintiles both had ratio values much lower than the national average, meaning they favored appropriations per student over student aid per student relative to the national average.
Individual state differences are certainly evident in the quintile groupings shown in Appendix F. This analysis was broad, intending to offer a general picture of the appropriation-to-aid mix and its effect on total average performance for three report card categories. Since individual state nuances must be analyzed within the context of the state, the next section uses New Mexico as an example of how the general analysis in the first three sections serves as a guide to offer ideas and recommendations concerning one state's performance.