Home Economics Economic Insights on Higher Education Policy in Ireland: Evidence from a Public System
The results are presented in two main sections: Sect. 5.4.2 examines variation in non-progression across the three main sectors: university, IT and other colleges (predominantly CEs), focusing on the role of composition (gender, age, social background, Leaving Certificate performance, grant receipt) and course type (field of study and course level) in non-progression rates in the three higher education sectors. This model is largely to illustrate the methodology adopted and the importance of taking account of student intake and course provision in measuring institutional effectiveness. Section 5.4.3 then examines variation in non-progression across all individual HEIs, presenting raw results and results adjusted for differences in student intake and course provision.
Characteristics of Students Who Do Not Progress
The first set of analysis, presented in Table 5.2, focuses on the chance of a student not progressing, with cumulative models taking into account:
The results show that, overall, males are less likely to progress from first to second year and are almost 1.4 times more likely to be in the nonprogression group than females. However, this gender differential predominantly reflects lower levels of (upper secondary) Leaving Certificate performance among male higher education entrants. When Leaving Certificate performance is taken into account, males are just 1.1 times more likely to be in the non-progression group. The gender difference is no longer significant when field of study and course level are controlled for, indicating that ceteris paribus, in terms of ‘ability’ and type of course taken, males are no less likely to progress than their female counterparts. This is consistent with the findings of McGuinness et al. (2012). Considering the age of students, mature students who are at least 23 years of age in the first year of their studies are significantly less likely to not progress than younger students.
Institutions varied considerably in the completeness of the data collected on the social class background of students. Using the information available (which is broadly representative of the population of new entrants) and including individuals where social class information is not provided in a separate category, strong social class differentiation in progression rates is evident. Students from manual and non-manual backgrounds are significantly less likely to progress than those from professional/managerial backgrounds. For the most part, social class differences in progression are largely mediated by Leaving Certificate
Table 5.2 Logit regression model of the factors associated with non-progression into second year
Table 5.2 (continued)
Notes: Results are presented as odds ratios. The reference categories are:
Female; Age 18 or younger; Professional/Managerial; 305-350 points; No grant; University; Social Science, Law and Arts; NQAI Level 8. t statistics in parentheses. *p < 0.05, **p < 0.01, ***p < 0.001 Source: Analysis of HEA data
performance—there are no longer significant social class differences in progression once Leaving Certificate performance is included (Model 2). While not shown here, the skilled manual group is the only exception— this group displays significantly lower progression rates than the semiskilled manual group. This may bear some relationship to the low (and declining) levels of student aid eligibility among this group (McCoy et al. 2010b) and the fact that this group are often on the margins of the student aid eligibility income thresholds (McCoy et al. 2010a).
This finding is reinforced by results showing significantly lower levels of non-progression among student aid recipients, a finding which remains even after taking account of the type of course taken and institutional sector. This indicates that financial support plays an important role in student retention—perhaps due to greater financial security, reduced reliance on (often difficult to secure) part-time work or simply students ensuring that they fulfil the requirements of their courses to retain student aid eligibility (since students who fail their exams and are required to repeat the year lose their eligibility for student aid). Internationally, research shows that financial support plays an important role in reducing dropout—see, for example, Lassibille and Gomez (2008) in the Spanish context and Dynarski (1999) and Bettinger (2004) in the US context. In the UK, Yorke (1998, p. 59) concluded “scholarships and grants tend to have the greatest beneficial effects on [college] persistence”. Additional analysis (not shown here) examined the extent to which the impact of student aid receipt varied across ‘ability’ groups: results showed that the impact of student aid receipt on progression chances was even greater for students with higher Maths performance levels.
As shown in Fig. 5.1, Leaving Certificate performance also plays a central role in student progression—the relationship is linear with rising points predicting lower non-progression, a finding which holds when taking account of field of study and course level. For each additional rise of 50 points, non-progression odds fall steadily: for example, relative to those securing 305-350 points, students who achieved 255-300 points are 1.2 times more likely to drop out, while those with 205-250 points are twice as likely to not progress to second year (Model 4). It is interesting to note that Leaving Certificate performance plays an equally important role in student retention in both the university and IT sectors,
Fig. 5.1 Overall non-progression odds by leaving certificate points. Source: Analysis of HEA data signalling the importance of student ‘ability’ in meeting the academic demands of higher education. The results also highlight the importance of academic preparedness prior to entry and adequate learning supports on entry to higher education. In the Spanish context, Lassibille and Gomez (2008), with similar results, argue that reducing the entry standards to satisfy the demand for higher education from an increasing pool of secondary-school leavers who are not necessarily equipped with the basic skills needed to succeed in higher education would have adverse effects. They argue that tighter selection at the point of entry to higher education might be needed. In the Irish context, given the numerus clau- sus system in operation, the academic requirements for entry reflect variation in student demand for courses and result in considerable variation between fields of study and institutions (and over time) in the academic ‘standard’ of higher education entrants. This makes it more difficult to impose higher education entry standards.
Additional models presented in Table 5.3 examine progression patterns according to Irish, English and Maths performance in the Leaving Certificate,7 rather than overall points achieved. The overall non-progression odds based on performance in these three core subjects are displayed in Fig. 5.2. In all three subjects, students with lower performance are more likely to not progress, while those with higher performance levels are significantly more likely to progress. It is interesting to note that the influence of Maths performance is greater than performance in English, while Irish performance is least likely to influence non-progression in higher education. Students with lower points in Maths are 1.7 times more likely to not progress to second year than are students with medium points (Table 5.3, Model 4). These findings indicate that students with low levels of performance in Leaving Certificate Maths struggle to meet the academic standards of college. However, additional analysis (not shown) examined to extent to which English and Maths performance was equally important in progression across all fields of study. The results point to Maths being significantly more important in student success in computer science, engineering and construction. Recent research in the University of Limerick (Treacy and Faulkner 2015) found that even controlling for student performance in secondary level Maths, the Maths skills of beginning undergraduates (in science and technology-based courses) was significantly below the performance of undergraduates ten years previously.
Table 5.3 Logit regression model of the factors associated with non-progression into second year: Irish, English and Maths attainment
Table 5.3 (continued)
Notes: Results are presented as odds ratios. The reference categories are:
Female; Age 18 or younger; Professional/Managerial; Irish moderate attainment; English moderate attainment; Maths moderate attainment; No grant; University; Social Science, Law and Arts; NQAI Level 8. t statistics in parentheses. *p < 0.05,
**p < 0.01, ***p < 0.001
Source: Analysis of HEA data
Fig. 5.2 Non-progression odds by leaving certificate performance in English, Irish and Maths. Source: Analysis of HEA data
|< Prev||CONTENTS||Next >|