Figure 1: Scheduled Hours Per Week by Subject Area

The Academic Experience of Students in English Universities

Tom Sastry and Bahram Bekhradnia

Higher Education Policy Institute

September 2007Introduction

In March 2006, with a grant generously provided by the Higher Education Academy, the Higher Education Policy Institute commissioned Opinionpanel Research to undertake a survey of first and second year students in English universities retained as panellists in their database. The survey focused on various aspects of the amount of teaching and private study undertaken by students and their levels of satisfaction and other attitudinal questions.

In October 2006, the report of the survey was published as The academic experience of students in English universities[1]. This is referred to in the text as ‘the 2006 report’. References to ‘2006’ should be taken as applying to the 2006 report or the survey on which it is based.

The survey was repeated in March 2007 again with the help of generous support from the Higher Education Academy. Thanks are also due to Opinionpanel who again agreed to conduct the survey at cost price.

Confirmation of 2006 findings

One purpose of repeating the survey was to validate the general account of the English Higher Education sector provided by its predecessor. The results here are very clear. The very impressive degree of consistency between the 2006 and 2007 survey enables us to say that the quantitative indicators of learning and teaching provision in English universities at whole system and subject level provide an accurate picture of provision in English universities. Whilst care needs to be taken at lower levels of aggregation, it is now possible to say definitively that:

Students in English universities typically receive around 14 hours of tuition per week (a weighted average of 14.2 hours in 2007, 13.7 in 2006)
Subject variations are both wide and consistent. Students in clinical and veterinary subjects typically receive just over 20 hours teaching per week; at the other extreme students in historical and philosophical studies typically receive between 8 and 9 hours.
The average student spends roughly 13 hours on private study (12.7 hours in 2007; 13.1 in 2006)
The total workload of English students averages around 25.5-26 hours (25.5 in 2007; 26.0 in 2006).
For students of medicine and dentistry, first and second year study is the equivalent of a full-time job at over 35 hours; for others it resembles part-time employment. Students of mass communications and documentation averaged 19.9 hours in the 2006 survey and 20.3 hours in 2007)
Students at old universities (Russell Group and pre-92 institutions) often receive most of their small group teaching from non-academics[2]. This pattern is not evident in newer universities.

These conclusions provide a basis on which to

Compare the English sector with other countries for which similar data have been collected
Place the quantitative data in the context of the conclusions of the academic literature on teaching and learning
Relate the results of the HEPI surveys to the findings of the National Student Survey – the definitive source of information on student satisfaction

Availability of raw data

Having confirmed the validity of the survey based approach to quantifying academic provision in England, HEPI is now releasing the raw data on which this report and its predecessor are based.

Comparisons between 2007 and 2006

In the 2006 survey, weightings were employed to prevent subject and year effects from biasing the results. In 2007, for the sake of simplicity, these have not been employed in quite the same way. Where it is most important to do so we have weighted for subject effects. There is no weighting for year effects in the 2007 results (i.e. to distinguish between first and second year students). The text and footnotes indicate where weightings have been used and highlights cases where the fact that they have not been used may be significant.

A further complication is that most of the questions have been modified slightly and some have been substantially redrafted. (The questionnaire, reproduced at Annex A, can be compared with the 2006 questionnaire available as Annex A to the 2006 report.) For these reasons it would not be appropriate in this report to provide a commentary on how English higher education has changed between 2006 and 2007 on the basis of a comparison between the two surveys – the changes observed are, in most cases, very slight and could have been caused either by random variation or changes in the approach or a combination of the two.

Notwithstanding the above, there are good grounds to regard the similarity between the 2006 and 2007 results as validating the general approach to the survey. The consistency between the 2006 and 2007 results suggests that students are able to recall details of their previous term’s work with sufficient accuracy to provide meaningful results. This point is discussed in detail in Annex B.

Hours of teaching

Students were offered a weighted mean of 14.2[3] hours of teaching per week. As in 2006, the highest levels of teaching were evident in health science and engineering subjects, the lowest in social science and the humanities. The three subjects with the lowest hours of teaching (historical and philosophical studies, linguistics and social studies) had less than half the level of teaching of the most heavily taught subject (veterinary and agricultural science).

Figure 1: Scheduled hours per week by subject area[4]

There is an impressive consistency between the 2007 results and those of the 2006 survey. Figure 2 shows the subject groupings used in the survey ranked in order of the mean number of scheduled hours in 2007. It is immediately apparent that the 2006 results painted a very similar picture.

Figure 2: Subject groupings by rank order of mean scheduled hours (highest = 1)[5]

In 2006, the survey found that the amount of teaching in old and new universities was broadly similar (13.7 hours in old and 13.3 in new). For the 2007 survey, the two categories have been split – old universities into Russell Group and non-Russell Group institutions and new universities into post 92 universities and other institutions. The raw average (mean) for Russell Group institutions at 15.2 is much higher than the mean for other pre-92 institutions and post-92 universities (both are at 13.5). However, this inevitably reflects the concentration of the subjects with the highest levels of teaching input – science and medicine – in Russell Group universities. The weighted figures given in Table 3 below give a truer idea of the size of the ‘type of institution’ effect.

Table 3: Mean scheduled hours of teaching by institution type[6]

Subject / All universities / Russell Group / Post 1992 / Pre 1992 (not Russell) / Other
Medicine and dentistry / 21.3 / 21.3 / 22.6 / 20.7 / n/a
Subjects allied to medicine / 18.8 / 19.3 / 18.6 / 19.3 / 15.7
Biological Sciences / 14.8 / 16.3 / 13.8 / 14.8 / 11.7
Veterinary agriculture and related / 22.2 / 26.4 / 14.6 / 14.6 / 27.9
Physical Sciences / 17.2 / 18.9 / 14.4 / 17.1 / n/a
Mathematical & Computer Sciences / 15.9 / 17.1 / 14.4 / 16.3 / 15.6
Engineering & technology / 19.3 / 20.4 / 16.4 / 20.2 / n/a
Architecture, Building & Planning / 16.4 / 16.1 / 16.5 / 16.6 / n/a
Social studies / 10.9 / 10.8 / 11.5 / 10.4 / 11.6
Law / 11.6 / 11.8 / 11.5 / 11.6 / n/a
Business & Administrative studies / 12.3 / 13.3 / 11.9 / 12.5 / 11.5
Mass Communications & Documentation / 12.0 / 11.8 / 12.3 / 12.2 / 9.6
Linguistics, Classics & related subjects / 10.2 / 10.8 / 10.2 / 9.8 / 9.0
Historical & Philosophical studies / 8.4 / 8.0 / 9.3 / 8.1 / 10.4
Creative Arts & Design / 13.2 / 10.7 / 14.0 / 12.4 / 13.5
Education / 13.6 / 9.5 / 13.9 / 11.2 / 14.3
All[7] / 14.2 / 14.4 / 13.7 / 14.0 / n/a

Unattended teaching

By subtracting the hours of teaching attended from the number of scheduled hours, it is possible to derive a measure of the proportion of teaching sessions not attended by students. In both 2006 and 2007, students reported non-attendance rates of less than 10 per cent.

It is to be expected that different subjects have different rates of non-attendance as this will reflect the extent to which all courses are mandatory. In fact, the range is not particularly wide with all subjects having reported non-attendance rates below 14 per cent.

While there may be little value in trying to identify trends from a comparison of 2006 and 2007 data, nevertheless, the consistency of the two years’ results suggests that the general pattern found in 2006 was accurate. Looking at the ranking of subjects as shown in Figure 4, the consistency between2006 and 2007 results is very striking. In 2006, the five subjects in which the highest proportion of scheduled teaching was not attended were computer science, business and administrative studies, social studies, mathematical science and law. In 2007 the same five subjects occupy the top four places (maths and computing have been merged for 2007, meaning that they occupy only one place between them). Perhaps unsurprisingly, as in 2006, education, veterinary science, medicine and subjects allied to medicine occupy the bottom four places. A plausible explanation for this is that the role of powerful licensing and/or commissioning authorities in curriculum design means that very little of what is taught is either superfluous or optional.

Figure 4: Percentage of scheduled hours of teaching not attended - by subject area

As Figure 5 shows, in most (82 per cent) institutions the mean proportion of unattended hours is between 4 per cent and 11 per cent.

Figure 5: Percentage of scheduled hours not attended[8] -by institution

Private study

In 2006 we reported that the weighted[9] mean amount of private study was 13.1 hours per week. In 2007 it was 12.5 hours.

Figure 6: Hours of Private Study by subject[10]

The data on hours of private study once again offers strong evidence that the survey approach provides good data at these levels of aggregation. If students were unable to estimate the occurrence of unstructured occasions (such as private study) with sufficient accuracy to enable surveys such as this one to provide useful information, we would expect to see considerable variation between 2006 and 2007 in the rank order of subjects. As Figure 7 shows, this has not happened.

Figure 7: Subject groupings by rank order of mean private study (highest =1)[11]

Effect of gender

As in 2006, there appears to be an association between gender and attendance and between gender and private study, as Table 8 shows.

Table 8: Private study and unattended hours of teaching by gender[12]

Hours of private study / Percentage of hours unattended
Male / 11.7 / 11.0
Female / 13.3 / 7.4

Total workload

In the light of what has already been said about the similarity in hours of teaching and private study between 2006 and 2007 it is unsurprising that Figure 9 (which shows total workload – attended hours plus private study) looks very similar to the equivalent chart in last year’s report with an overall weighted mean of 26.0 hours compared to a weighted mean of 25.7 hours in 2006 and with scientific and health related subjects, together with architecture, showing the highest workloads.

Figure 9: Student workloads: hours of teaching plus private study – by subject[13]

Students in Russell Group universities spent more time on average on their studies than others[14]. Their mean workload (attended hours of teaching plus private study) was 28.2 hours compared to 24.5 for other pre-1992 institutions and 24.1 for post 1992 universities. These figures are of course, influenced by the subject mix – medical and scientific subjects, which have high workloads, are concentrated in Russell Group institutions. Even the weighted means shown in Table 10 below, however, show evidence of a small ‘Russell Group effect’ albeit a less dramatic one than the raw figures suggest.

Table 10: Total workload (hours) by subject and type of institution[15]

Subject / All universities / Russell Group / Post 1992 / Pre 1992 / Other
Medicine and dentistry / 35.9 / 36.1 / 33.5 / 36.0 / n/a
Subjects allied to medicine / 30.4 / 30.2 / 31.2 / 29.6 / 28.1
Biological Sciences / 25.0 / 26.7 / 23.8 / 25.0 / 22.6
Veterinary agriculture and related / 33.7 / 37.7 / 24.0 / 24.4 / 41.6[16]
Physical Sciences / 28.0 / 30.3 / 24.5 / 27.4 / n/a
Mathematical & Computer Sciences / 26.0 / 28.6 / 23.3 / 26.0 / 20.9
Engineering & technology / 29.2 / 30.2 / 26.9 / 29.6 / n/a
Architecture, Building & Planning / 31.1 / 33.3 / 29.8 / 31.3 / n/a
Social studies / 22.0 / 23.7 / 21.8 / 21.0 / 22.3
Law / 26.5 / 31.4 / 23.2 / 25.4 / n/a
Business & Administrative studies / 20.9 / 22.6 / 20.1 / 21.3 / 21.8
Mass Communications & Documentation / 20.3 / 20.1 / 20.7 / 20.2 / 17.4
Linguistics, Classics & related subjects / 23.2 / 25.0 / 21.7 / 22.8 / 19.0
Historical & Philosophical studies / 22.5 / 24.7 / 19.8 / 21.1 / 24.6
Creative Arts & Design / 25.2 / 24.4 / 26.0 / 23.0 / 24.0
Education / 25.3 / 21.4 / 25.5 / 22.8 / 26.7
All subjects[17] / 26.0 / 26.7 / 24.3 / 24.8 / n/a

The findings outlined in the previous paragraph should not be over-stated. As in 2006, the variation between individual institutions is very much greater than the variation between types of institution, suggesting that the differences between universities of the same type are at least as important as the differences between types of institution. Table 11 shows this very clearly, and summarises the information at Annex E which contains tables that show for each subject the average number of hours of total workload in each institution.

Annex E also shows the number of “good” (2:1 and above) degrees awarded, by subject and institution, along with the average number of UCAS tariff points of their entrants. It is clear from this that in some subjects and in some universities it is much more difficult to obtain a good degree than in others – students need to have better entry qualifications and work harder.

Last year’s report observed that “In particular it raises questions about what it means to have a degree from an English university, if a degree can apparently be obtained with such very different levels of effort. Some institutions award many more 2.1 and first-class degrees than others, and there are subject differences too. Explanations for this might be that the students concerned are more able, or else that they work harder… On the basis of these data, neither of these explanations appears to provide a complete answer”. That observation remains true.

Others have pointed out that the degree classification system does not provide a basis for comparing degree standards, and this report adds weight to that conclusion: it certainly raises questions that need to be addressed. Since last year’s report, the Burgess Committee has completed its work, and is expected also to conclude that the degree classification system is no longer fit for purpose, but that identifying an acceptable alternative is a challenge. While these data certainly do not prove that the degree classification system is flawed, they nevertheless do raise questions that need to be addressed[18].

Table 11: Student workload by subject – highest and lowest institutional mean hours per week (average of 2006 and 2007)

Subject / Highest institutional mean / Lowest institutional mean / Median
Medicine and dentistry / 46.3 / 26.3 / 35.5
Subjects allied to medicine / 38.3 / 24.6 / 31.2
Biological Sciences / 39.9 / 15.0 / 24.5
Veterinary agriculture and related / 41.6 / 23.5 / 37.0
Physical Sciences / 45.3 / 19.8 / 27.6
Mathematical & Computer Sciences[19] / 36.4 / 17.1 / 26.2
Engineering & technology / 41.2 / 20.8 / 28.7
Architecture, Building & Planning / 41.5 / 26.3 / 28.5
Social studies / 35.8 / 14.0 / 21.6
Law / 44.8 / 18.7 / 26.2
Business & Administrative studies / 28.3 / 15.5 / 20.8
Mass Communications & Documentation / 26.8 / 14.7 / 19.4
Linguistics, Classics & related subjects / 39.3 / 14.8 / 22.3
Historical & Philosophical studies / 39.5 / 14.0 / 21.5
Creative Arts & Design / 34.5 / 17.2 / 25.6
Education / 33.7 / 14.4 / 25.5

Size of teaching groups

In 2006, we reported that students received a mean of 3.5 hours of teaching in small groups (with up to fifteen other students). As Figure 12 shows the 2007 results are very similar. The overall mean is 3.6 hours.

Figure 12: Amount of teaching in groups with 15 or fewer other students (in addition to the respondent) by subject area[20]

As Figure 13 shows, when subject groupings are ranked on the basis of the amount of small group teaching the 2007 result is very similar to the 2006 result.

Figure 13: Subject groupings by rank order of mean hours in groups with 0-15 other students beside the respondent (highest =1)[21]

In 2006, students at old and new universities reported similar amounts of very small group teaching (with 0-5 others) whilst new universities appeared to provide much more teaching in the 6-15 range (4.1 hours as opposed to 3.2 for old universities). This pattern is replicated in 2007.

Table 14: Mean number of hours in small group sessions – old and new universities[22]

0-5 others / 6-15 others / 0-15 others
All institutions / 0.8 / 2.8 / 3.6
Russell Group / 1.0 / 2.3 / 3.4
Other pre 92 / 0.5 / 2.5 / 3.0
Post 92 / 0.8 / 3.4 / 4.2
Other / 0.6 / 3.1 / 3.7

Use of specialist academic facilities

In 2006, respondents were asked about supervised and unsupervised use of specialist facilities. For the 2007 survey, the questions were changed: students were instead asked to report how much of their teaching hours and private study involved the use of specialist facilities. The results are shown in Table 15 below.

Table 15: Use of specialist facilities in taught sessions and private study by subject

Teaching / Private study / Total
Medicine and dentistry / 6.2 / 1.3 / 7.5
Subjects allied to medicine / 5.5 / 1.1 / 6.6
Biological Sciences / 4.5 / 1.1 / 5.6
Veterinary agriculture and related / 6.5 / 0.9 / 7.4
Physical Sciences / 5.5 / 1.1 / 6.6
Mathematical & Computer Sciences / 3.1 / 2 / 5.1
Engineering & technology / 5.1 / 2 / 7.1
Architecture, Building & Planning / 5.9 / 4 / 9.9
Social studies / 1.2 / 0.9 / 2.1
Law / 1 / 1.3 / 2.3
Business & Administrative studies / 1.5 / 1.2 / 2.7
Mass Communications & Documentation / 3.2 / 1.8 / 5
Linguistics, Classics & related subjects / 1 / 1 / 2
Historical & Philosophical studies / 0.6 / 0.8 / 1.4
Creative Arts & Design / 5.8 / 3.4 / 9.2
Education / 2.5 / 1.2 / 3.7
All subjects (weighted) / 3.4 / 1.5 / 4.9

Teaching led by non-academics

One of the most striking findings of the 2006 survey was that 30 per cent of students in old universities reported that seminars and tutorials were led mainly by non-academic members of staff (the figures for new universities were much lower at 8 per cent for seminars and 7 per cent for tutorials).

Figure 16 shows that the general pattern is unchanged: students at Russell Group and pre-92 universities report much higher rates of teaching by non-academics, particularly where teaching groups are smaller[23].

Figure 16: Percentage of respondents[24] reporting that seminars and tutorials were led mainly by non-academics

Paid work

Figure 17 shows that students who do more hours of paid work not connected with their courses tend to perceive poorer value for money than those who do less. This finding was also noted in 2006. In neither year was the effect a particularly strong one.

Figure 17: The impact of paid work on value perception: percentage reporting poor value for money by hours of paid employment (numbers of responses in brackets)

Informal tuition – discussions with staff outside scheduled teaching

In 2006 the survey asked about the frequency of substantive discussions with staff outside scheduled hours of teaching. For 2007, the survey shifted from using ordinal categories (‘less than once a month’, ‘once a month’ etc.) to asking students to estimate the number of unscheduled contacts. This gives us for the first time a measure of the amount of contact students had with staff.

As Figure 18 shows, the mean of 1.8 contacts is quite substantial[25]. Assuming a ten week term this equates to 0.2 contacts per week. If each contact lasted half an hour and is on a one-to-one basis, this is equivalent in terms of staff time to an additional 10-person seminar each week – or ten 100-person lectures. Unscheduled contacts are likely to be highly skewed because they depend on the willingness of students to seek and obtain the attention of staff. It is probable, therefore that there is a minority of students for whom unscheduled contact adds very substantially to the amount of staff time invested in their teaching. This potentially raises issues of equity – it may be that a minority of more assertive students are gaining a considerable advantage through this form of informal tuition.

Figure 18: Mean substantive unscheduled contacts with academics in previous term or semester[26]