Tutors and the student transition into Higher Education: Supportive or survivalist?

John Buglear

Nottingham Trent University

Abstract

The role of tutors in student transition and retention might be self-evident, but what attitudes shape their approach to this role? This paper reports the results of a survey of academic staff designed to measure the degree of supportive disposition among them. The results from nearly 300 respondents suggest that tutors are generally supportive. The survey also revealed evidence of variation associated with gender, role, and length of service. A comparison with results of a pilot investigation indicates that attitudes fluctuate during the academic year.

Introduction

The origins of the work presented in this paper are rooted in the author’s experience of managing large business courses in Nottingham Business School and the concomitant focus on student progression. Perhaps reflecting the diffuse perceptions and expectations of business that they appeared to hold, recruits to these courses seemed to have a greater propensity to ‘drop out’ than was the case with courses in other disciplines. During this period of a decade or so the prominence of student retention, the effecting of successful transition into higher education, grew in the public domain, with one Secretary of State urging institutions to ‘bear down’ on dropping out (Blunkett, quoted in Yorke and Longden, 2004, 50). Since David Blunkett’s admonition in 2000 the national profile of the non-completion issue has been maintained (National Audit Office, 2007; House of Commons Committee of Public Accounts, 2008).

There is considerable evidence that interaction with tutors is a key attitude-forming aspect of students’ early experiences of higher education. From their investigation of the experiences of accounting undergraduates Gracia and Jenkins concluded that

‘Most students highlight the importance of the tutor as central contributors to their ‘best’ and ‘worst’ learning experiences.’ (2002, 101)

According to Bryson and Hand

‘The considerable research that has been conducted upon students’ approaches to learning carries an encouraging message: that approaches to learning are not fixed within an individual, but […] can be influenced by the teacher.’ (2005, 104)

The authors of the two articles from which the above quotes are taken are themselves academics, as are the authors of other works reaching similar opinion about the significance of tutors (Baker, 2006; Cartney and Rowe, 2006; Foster, 2005; Rhodes and Neville, 2004; Wilcox et al., 2005). Whereas it might be argued that academics are unlikely to underplay their own significance UNITE, according to its own website ‘the UK’s leading developer and manager of student accommodation’ (UNITE, 2009), might be deemed a more independent source. From their survey of 1600 students across twenty UK universities they concluded that

‘What is clear is that in the first instance academic factors are significantly more important than any facilities that the university provides or the social environment of the towns/cities where they are located.’ (UNITE, 2007, 11).

For some students interaction with arguably the key academic factors, tutors, can make the difference between persevering with their course of study and leaving (Fitzgibbon and Prior, 2003). The stance of tutors towards such students is therefore of interest to anyone concerned to promote successful transition into higher education and thereby improve student retention. In view of this the following research questions were posed:

1.  What are the perspectives on non-completion (‘dropping out’) held by Nottingham Trent University (NTU) academic staff?

2.  How might these perspectives be measured?

3.  Are these perspectives related to demographic and occupational factors e.g. gender, length of service?

Research methods

Anderson postulated a dichotomy of stances on student retention; he contrasted the survivalist to the supportive or remedialist. The former might be characterised as laissez faire, ‘sink or swim’ academic Darwinism based on the survival of the intellectually fittest and hence the expectation that ‘students will be self-motivated and self starting’, whereas the latter is epitomised by a commitment to the delivery of remedial support (Johnston and Simpson, 2006, 29). The measurement of academic staff perspectives applied in this study is based on an attempt to calibrate Anderson’s dichotomy.

The calibration method is founded Thurstone’s ‘method of equal appearing intervals’ (1929, 165) that was developed to measure ‘psychological value’ or attitude (Thurstone, 1929, 157). Oppenheim deemed the method to be especially appropriate when group differences are of interest (1992, 189), as is the case with the third research question addressed here. The validity of the method has been demonstrated although the viability of benchmarking the extremes of a Thurstone scale using individuals with verifiably extreme attitudes is open to question (Oppenheim, 1992, 194/5).

The construction of a Thurstone scale begins with the generation of attitude statements (Oppenheim, 1992, 179). For this study almost one hundred statements were produced, reflecting a range of sentiments towards students who drop out. Two examples are:

‘I think that students who drop out tend to be spoilt brats who can’t make it without Mum and Dad holding their hand.’

‘I want all students who come here to succeed.’

The set of statements were circulated to a panel of nineteen judges who rated each statement on a scale of one to eleven, where one represented extreme survivalism and eleven, extreme supportivism. For each statement the median and inter-quartile range of the judges’ scores were calculated. The first of the examples above had a median judge score of one, the second a median score of eleven.

Using the median judge scores a list of eleven statements, one for each point on the putative survivalist-supportivist scale of one to eleven, was compiled. In cases where several statements in the original list had the same median the one with the least spread of judges’ scores was selected.

The eleven statements, presented in randomised order, were the basis of a questionnaire. For each statement the median of the judges’ scores was used as the attitude score of the statement, although this was not shown on the instrument. Questions on demographic and occupational factors were added.

Respondents were asked to indicate agreement or disagreement with each of the eleven statements. The attitude score of the respondent was the median of the attitude scores of the statements with which the respondent agreed. To illustrate, a respondent who agreed with the statements scored 3, 4, 6, 7, and 8 would be deemed to have an attitude score of six. The lower the attitude score, the more survivalist the respondent, the higher the attitude score the more supportive the respondent.

The questionnaire was distributed by email to over 1000 academic staff at NTU using the HR Information System lists. Since it appeared that part-time temporary staff were largely absent from these lists a search of the email system using the string ‘part-time sessional hourly-paid’ was used to identify email addresses for such staff. This yielded a list of several hundred addresses, although a substantial proportion of them were later found, for a variety of reasons, to be inaccessible.

The questionnaire was launched in late February 2007 with follow-up requests during the following month. The responses were delivered to an independent third party to preserve the anonymity of respondents.

Findings

By the time the survey was closed almost 300 responses had been received. This constituted a response rate of approximately 25%. There were only three responses from the list of temporary part-time staff. The response rate based solely on the HR lists was approximately 30%. The data were used to assess the extent to which attitudes were associated with, inter alia gender, school, seniority, holding course and module leadership roles, balance of workload and work satisfaction. The data from one school, Business, was compared to the results from a pilot study conducted in the school to explore whether attitudes are consistent over the academic year.

The response rate varied by school, as illustrated in Figure 1. The highest response rate, nearly 50% was from Business School. This might reflect the school being where the author works, his using the school in a pilot study some months before, and a sensitivity to what some staff in the school felt were relatively modest progression rates. The lowest response rate was from Agriculture, which could reflect the school being based at a relatively remote campus and, by virtue of their being relatively recently incorporated into NTU, lower staff affinity with the university.

The overall distribution of the 295 respondents’ scores is depicted in Figure 2. The values are concentrated around eight and nine, implying that in general respondents were relatively supportive. There is a slight negative skew, reflecting the few relatively survivalist respondents. There was no discernible consistency in their characteristics, save that the majority were male.

Figure 3 shows a comparative boxplot1 of scores by gender. The slightly higher median for the 149 female respondents suggests they are on average more supportive. The greater number of asterisks below the box on the right demonstrates that more of the lower outliers are among the 146 males.

The comparative boxplot in Figure 4 portrays the distribution of respondents’ attitude scores by school. The highest median score is for the fourteen Agriculture respondents, the lowest for the sixty Business respondents. These schools generated the lowest and highest response rates respectively. Conceivably the Agriculture median might be higher because of a preponderance of survivalist attitudes among the non-respondents. The largest spread is among the thirteen respondents from Computing, the least among the fourteen Agriculture and 44 Social Science respondents.

The diagram in Figure 5 shows the attitude scores by post held by the respondent. The nine academic managers (Deans and Division Heads) had the highest average attitude scores, with the nineteen Readers and Professors and the 224 main grade lecturers providing jointly the lowest median. The largest spread of attitude scores was among the main grade lecturers.

Figure 6 presents the attitude scores by whether or not the respondent was a course manager. The distributions are similar, although there are several outliers among the 145 respondents who were not course managers compared to only one among the 149 who were course managers. These results constitute little evidence to support the assertion that course managers are more supportive, which might be supposed by virtue of having a specific concern for the progression of the students on their courses.

The boxplots in Figure 7 portray the attitude scores of module leaders compared to those with no module leadership role. The 242 module leaders have on average lower scores that are more widely spread than the scores of the 52 respondents who were not module leaders. There is also a set of lower outlying values from the module leaders. The results might reflect module leaders having responsibility for setting and marking referral for less engaged students.

The hypothesis that constituted the rationale for the analysis depicted in Figure 8 was that those with the lowest proportion of teaching in their workload were likely to be subject rather than student orientated and hence less supportive. The results, with slightly lower medians for those with the lowest and highest proportion of teaching in their workload (44 and 109 respondents respectively), indicate that on average both of these categories of staff are less supportive than their colleagues with more mixed workloads (24 respondents were in the second category and 116 in the third).

Figure 9 shows the distributions of scores by categories of response to a question about job satisfaction (‘Generally how satisfied are you with your current job?’). The fifty respondents who were neither satisfied nor dissatisfied have on average slightly more supportive attitudes than the 86 who expressed dissatisfaction and the 157 who expressed satisfaction, but there is little difference between the distributions in terms of spread. The notion underpinning this analysis, that job dissatisfaction is associated with less supportive dispositions is not supported by this evidence. The negative sentiments dissatisfied respondents might harbour towards their employer seem not to be manifested in distinctly different attitudes towards their students.

Figure 10 shows the scatter of attitude scores and length of service in higher education. There is low but marginally significant negative correlation between the two variables (r = -0.104, P = 0.08), putting the hypothesis of no association between attitude score and length of service open to question. This implies that with longer service comes a less supportive outlook, or possibly that those who now have long service entered the sector at times when expectations about students were very different, and that to some extent they retain attitudes from those times.

The results presented hitherto are from the survey of academic staff undertaken in the spring of 2007. A pilot study of Business School staff was carried out in July 2006. Figure 11 shows the distribution of attitude scores from this pilot study and the attitude scores of Business School staff from March 2007. The results show a distinctly lower average score in July, and the difference is statistically significant. It might be conjectured from this that the lower supportive dispositions in July reflect the focus on examination boards and ensuing referral tasks arising from the shortcomings of students who might too easily be deemed indolent. The difference should however be considered in the light of the facts that firstly the instrument used in the pilot was similar to but not exactly the same as that used in the survey, secondly that the Business School scores are not paired2, and thirdly that the response rate in the pilot study was considerably higher at 75% than the 45% achieved from the Business School in the main survey.

As has been noted earlier the response rate was lower in some schools than others. It is open to question how non-response bias could have influenced the results. Perhaps those who did not respond were less supportive and thus were simply disinclined to participate in an investigation about student retention. There is of course no way of knowing what non-respondents’ attitude scores would have been but according to Moser and Kalton