Copyright Sociological Research Online, 2000

 

Steve Fenton, John Carter and Tariq Modood (2000) 'Ethnicity and Academia: Closure Models, Racism Models and Market Models'
Sociological Research Online, vol. 5, no. 2, <http://www.socresonline.org.uk/5/2/fenton.html>

To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 11/5/2000      Accepted: 4/9/2000      Published: 6/9/2000

Abstract

The paper addresses racism, discrimination, equal opportunities policies, institutional cultures, and the pressures of markets in influencing the position of minority ethnic groups in academia. The representation and position of minority ethnic groups among academic staff in UK higher education has previously been little studied. Data from the Higher Education Statistical Agency records and from new surveys are presented and analysed. Representation is low especially among some groups, but is growing among younger sections of academic staff, and is much higher in some academic subject areas than others. Analysis of terms of contract and of seniority by ethnic groups suggests that minorities are significantly less well placed within the profession. An important distinction is between British and non- British nationality in assessing ethnicity and academic posts; non- British staff may be seen as part of a global labour market, especially in fixed term contract research work. The evidence is evaluated alongside a re-exploration of principal models for explaining ethnic disadvantage in labour markets: closure, discrimination, equal opportunities, institutional racism and markets. The authors conclude that a combination of the last two models offers the best prospect of a full explanation.

Keywords:
Ethnicity; Higher Education; Employment; Equal Opportunities; Racism; Academic Marketplace; Universities in UK; Discrimination, Institutional Racism

Introduction

1.1
Colleges and universities in Britain were originally part of religious institutions and their single-denominational foundations have meant that entry to the academic profession has been restricted by confessional exclusions (McClelland 1973) as well as by gender and by class advantages and class-based subcultures (Halsey 1992). These religious and gender discriminations were not incidental nor external to the logic of the institution, ideas of gentle breeding and social class too have been closely bound to ideas of learning; 'a gentleman and a scholar', we say, and scholars are expected to be gentlemen. English - much more than Irish or Scottish -universities themselves had, on the whole, a narrow social compass in staff and students until the 'opening up' of higher education in the 1960s (Halsey 1992). The expansion of entry of students to Higher Education in the post war period has led to some 'democratisation' of the profession (Stewart 1989; Halsey 1992) and the second expansion of the 1980s and 1990s has created in Britain something closer to the American system of mass higher education. Another kind of 'democratisation' occurred in 1992 with the granting of University name and status to institutions previously styled as Polytechnics. Although this expansion of higher education has not been accompanied by a matching allocation of resource per student, there can be little doubt that the social ambience of British Higher Education has changed. The older culture of higher learning and social exclusiveness persists but it sits alongside environments which differ greatly in architecture and temper. There are, thus, many and varied institutional cultures in British higher education today. Exclusion, discrimination, privileged and disprivileged entry and the cultural climate of British Higher Education are the themes of our discussions below - but in this paper the focus is on race and ethnicity at the turn of the millennium.

Ethnicity and University

2.1
Despite the intensified interest in ethnicity within academic sociology and a raised consciousness about 'institutional racism' on the political scene, there has to date been very little system-wide attention to ethnicity and the social composition of academics in British Universities (but see Modood T. and Acland T. eds 1998, Singh G. 1998 and Cohen P. 1995). New ground was broken when a project was funded[1] to carry out two surveys[2], to conduct some discussion groups, and to examine the Higher Education Statistics Agency (HESA) data set, containing information on over 126,000 academic staff in the UK in 1996-97[3]. Whilst the main findings and recommendations are published in the research report (Carter, Fenton and Modood 1999), in this paper we look more closely at some key findings, interrogating them in the light of sociological theorisations of ethnicity, culture, labour markets and disadvantage. The first part of the paper will outline the principal theoretical approaches to ethnic disadvantage, especially within labour markets; the second part will examine the evidence of the HESA data and our survey data; and the third part will draw what conclusions we can about the applicability of competing theories to ethnic minority staff participation in the academic profession.

Closure, Exclusion, Disadvantage

3.1
There are two related dimensions of ethnicity and the academic profession that we shall consider at the outset: the question of entry into the profession and the question of experience within it; we look first at entry. A group may be wholly or largely excluded from a profession, occupation or area of economic activity because those who dominate that area and control entry to it seek to protect themselves from competition or from challenges to their institutional power. Such exclusions may be explicit or tacit: historical exclusions of religious communities and of women from occupational zones have often been explicit as have racial exclusions in the United States, South Africa and many other countries. This has been described by Max Weber and by more recent class and labour market theorists as social closure, 'the process by which social collectivities seek to maximise rewards by restricting access to a limited number of eligibles' (Parkin 1974).

3.2
The emphasis in Parkin's formulation is on rewards - where scarcity is artificially preserved, those who are included can maintain higher rewards. But we may want to consider the plain wish, inspired by hostility to outsiders, to preserve a monopoly or dominance in control of an occupational sphere by excluding newcomers or making entry difficult. The ideological and institutional means whereby the insiders achieve this are varied. Common among them are ideologies proclaiming incumbents as more suited to the task (such as men as doctors, Stacey 1990), the excluded as unsuited (historically at least, women as doctors, Stacey 1990) or portraying outsiders as unreliable or even dangerous. The commonest institutional means are categorical exclusions, such as of Jews from occupational areas through much of European history, or the Irish and Roman Catholics in much of English history; and credential exclusions which make high credentialist demands and limit access to the credentials.

Gatekeepers' Judgements

4.1
Racial discrimination is a narrower idea than the idea of social closure, although without doubt discrimination 'on racial grounds' is a common form of closure, and discrimination by gatekeepers a common method by which closure is achieved. In USA labour markets white workers succeeded in excluding black workers from whole tracts of employment by violence, by tacit agreements with employers, or by exclusion of black workers from labour unions (Bonacich 1976); similar, if less overt, patterns can be found in England (Ballard 1994). The concept of racial discrimination does not always emphasise the gains which the 'included' may make; racial discrimination is often seen as the exercise of a conscious or unconscious 'irrationality', meaning the harbouring of prejudice against an outsider group in general. Gatekeepers stand at strategic points where they are capable of making discriminatory decisions, whether these are about entry to skilled manual occupations, medical schools, about allocation of houses, or memberships of private clubs. Gatekeeper discrimination is the type of closure made unlawful (in Britain) by the 1976 Race Relations Act. This act also provides for indirect discrimination - where qualifications required for a post tended to exclude a specific group and could not be justified as a genuine requirement - but often would not capture the broader kind of credentialist exclusion we have described above. Credentials are regarded as neutral and meritocratic but the truth is more complex and for three reasons: access to the credentials may be narrow; credentials have a formal equivalence whilst being informally ranked, and the assessment of credentials may be compounded with consideration of social fit. As Jenkins has observed, suitability criteria are overlain with acceptability criteria (Jenkins, 1986). The unspoken or disguised use of acceptability criteria probably accounts for much of the discrimination along gender, class and ethnic contours; it is difficult to trace and different in kind from 'they shall not pass' exclusion as practised by, say, white American workers or homeowners.

Institutional Racism

5.1
The concept of institutional racism has a much broader ambit than that of gatekeeper discrimination, incorporating ideas about a) institutional culture, b) routine practices, and c) social inequalities outside the institution. The institutional culture may live in the 'canteen' or the 'boardroom' or both, but it is racist if it constitutes a climate of assumptions which are hostile to outsider groups, racially or ethnically defined. It is not just the existence of such a culture that is critical but the fact that it goes unchecked. 'Routine practices' would include entrenched methods of recruitment, for example by word of mouth or sponsorship; and any standard practice which acts to the disadvantage of specific groups.

5.2
'Culture' and 'routine practices' are posited here as dimensions of the 'institution' in a largely substantive sense, describing the way an institution operates and its climate of attitude; its mode and its mood. The third dimension recalls the original Hamilton and Carmichael (1969) sense of institutional racism, meaning racism that permeated the ideological and material fabric of a whole social order. This consideration of inequalities outside the institution - that people may have been denied access to education which leads to credentials for example - takes us outside 'the institution' in the narrow substantive sense, thus being quite distant from a narrower contemporary policy-oriented usage in Britain, that has been used in reference to the Lawrence inquiry (Macpherson, 1999; cf. Solomos and Back 1996).

Equal Opportunities

6.1
The concept of institutional racism in its 'routine practices' version, clearly overlaps with a broader concept of discrimination (that is, a concept which incorporates the idea of indirect discrimination); in its 'external inequalities' version it overlaps with the concept of disadvantage and a racialised class structure. In its emphasis on institutional culture it is directly relevant to the pursuit of equal opportunities. This may be seen as the obverse of racial discrimination; but an equal opportunities argument is not merely that. Equal opportunities as a 'philosophy' advances the specific contention that carefully designed procedures are needed to ensure fair treatment. Given the existence of a culture and a set of routine practices which have not been reflected upon, the equal opportunities model forces reflection on the institutional culture and advocates change of the routine practices of hiring and promoting. Whilst a liberal approach to equal opportunities aims at removing any obstacles to the rewarding of merit, a radical approach recognises and addresses historical disadvantage (Jewson N and Mason D 1986).

Market Models

7.1
The final model of ethnic or racial disadvantage to consider is the market model presented here as having two 'versions': the first as market disadvantage, the second as market choice. Market disadvantage can be instanced by reference to Wilson's landmark work The Declining Significance of Race (Wilson 1980) in which he argued that whilst the exclusion and disadvantage of black Americans could be traced historically to racism and discrimination, their (then) current position was also traceable to the collapse of the labour markets in which they had typically worked and their lack of educational credentials for the new labour markets; and all of this in a context of class inequalities in which the narrowing of inequalities was on no-one's political agenda. If we are looking at professional occupations, we must examine class and educational advantages since these will play such a key part in determining entry to professional occupations. And market disadvantage has a clear conceptual overlap with the 'societal disadvantage' framing of institutional racism, even if the former 'backgrounds' racism and the latter 'foregrounds' it.

7.2
The idea of market choice directs our attention to the strategic choices, intentions and preferences of 'minority' group individuals themselves; they may face constraints but they also assert choices and chart their own paths (Dahya, 1974; Banton, 1983). For example, some groups may seek academic success more than others or may invest more in acquiring qualifications; once they begin to achieve a certain success in some academic fields, this may draw them further into those fields. Of course occupational outcomes will depend upon much else than such choices; they will depend upon factors such as the emergence of or decline of certain economic opportunities. But individual choices, structured by family, community and ethnic group efforts and values will be a contributing factor. Indeed if we look back over the models of closure, gatekeeper discrimination, institutional racism, equal opportunities, market disadvantage and market choice all but the last refer to actions and attitudes which are structured and/or controlled by the 'included'. The last is the only one of all these models which sees the 'excluded' actors as expressing cultural or strategic preferences. Action is always in a context of choice and constraint; little can be gained from arguments about the a priori merits of a choice or constraint model. Frequently, clearer answers will come from empirical enquiry. Three examples relevant to minorities illustrate this: the move into self- employment and small business by migrant and minority groups, patterns of residence and home ownership, and the purchase of private education. Some groups choose small business as a route to success, home purchase as a mark of independence, and private education as an investment; equally all three may be constrained by discrimination in the sense of being guided by the wish to avoid actual or anticipated obstacles.

Ethnicity and Academic Staff in UK Universities and Colleges

8.1
Since the HESA data uses the same ethnic code as the UK census, it shares its weaknesses, above all that approximately 94 per cent of the population are recorded as undifferentiated 'white', a colour and not an ethnic category (Fenton 1996). About one fifth of the staff recorded by the 1996 HESA data do not have information on ethnic group[4]; the data analysis below is almost all based on that 77 per cent of the data base for whom ethnicity and nationality is known, 97,533 UK academic staff. 5,335 of these are recorded as other than white. Both are contentious terms but 'ethnic minority' staff in UK universities is effectively 'non-white' staff. This is neither a desirable definition nor a definition consistent with current discussions of ethnicity. But it is the only one available to us from the collected data.

8.2
The present non-white British population largely derives from those migrating to Britain from the Caribbean, South Asia and East and West Africa since the mid-1950s. Over half of that current population is born in Britain and has a younger age profile than the UK population as a whole (Warnes, 1996; Modood et al, 1997). Its class profile over much of that fifty year period has shown concentrations in skilled, semi-skilled and unskilled manual and service occupations; in the 1990s the evidence has shifted towards showing greater representation in professional and other higher class groupings, and considerable difference between the profiles of specific groups (Modood et al. 1997) The present study examines whether that post-war non-white Britain-settled and Britain-born population has entered the academic profession and examines their position within it.

The Global Academic Marketplace.

9.1
There is a second population which must also be examined within the framework of this study. The academic profession, like other professions, attracts entrants in the UK from a global market, in academic training, careers and career development. Non-British academics come as doctoral students and post-doctoral researchers and will continue their careers both outside and within the UK. Their representation within the academic profession is not a representation of the Britain-born or the long-term settled non-whites or of the latter's social mobility within the UK; it is an index of the globalisation of the academic marketplace. The only available measure of this distinction in the HESA data is the recording of nationality, an imperfect[5] measure but a useful approximation to the distinction we have made. A combination of the categories 'white' and 'non-white' in ethnic groups, and British and non-British in nationality produce four groupings, white and non-white British and white and non-white non-British. 5.4 per cent academic staff are recorded as non-white, and among them the non-British outnumber the British; the 'global labour market' non-white academics are a greater presence than the 'British and non-white' academics'. A further 10 per cent of UK academics are white and non-British. (That is among those for whom ethnicity and nationality is recorded; there are a further group, more than 5,000, of non-British nationality for whom ethnicity is not recorded). Table 1 shows, for UK institutions 1996-7 (the data source referred to hereafter as 'HESA 1996'), all minority (non-white) academic staff by ethnic group and shows British-nationality minority groups as proportions of all British-nationality academic staff, comparing this with the representation of the same groups in the UK Census 1991 population.


Table1: Distribution of academic staff, by ethnic group The UK census data is given for comparison
Ethnic GroupAll Academic Staff*British Nationality Academic Staff*UK Population**
n%n%%
White92,19894.582,91197.295.1
Non-white5,3355.52,4182.84.9
Chinese1,6731.73400.40.3
Indian1,3901.48251.01.5
Pakistani2910.32190.20.8
Bangladeshi750.1440.10.3
Asian (other)8730.94230.50.4
Black African5560.61980.20.4
Black Caribbean / Black (other)4770.53690.41.2
Total97,53310085,329100100
* Data source: HESA, 1996
** Data source: UK Census, 1991

9.2
Some groups are notably over-represented as a whole (Chinese and Asian other in particular) but it is also clear that this over-representation is a consequence of the presence of non-British non-whites. On a simple comparison of British nationality minorities in the academic profession with minorities in the British population, all groups except Chinese and Asian Other are under-represented and Pakistani, Bangladeshi and Black Caribbean groups are notably so. It should be remembered that the Census does not contain nationality data and the populations by ethnic group contain both British and non-British nationals.

9.3
Some older British non-white academics have been in academic life for some time (21 per cent of non-white British staff have been in their present institution for 9 years or more compared to 39 per cent of whites). Numbers and proportions of minority staff, are greater in those with shorter duration of service (61 per cent of non-whites have been in their present post for 4 years or less compared to 42 per cent of whites). Equally the non-white academic staff is, as we have seen, a much younger population than the white. Minorities are younger staff and have been in their present posts for a shorter period. The better test[6] of representativeness is the proportions of minority and majority staff in the younger academic labour force. Table 2 shows that the representation of British non-white groups in academic staff is considerably greater among staff 44 years and younger compared with their total representation as seen in table 1. Comparing them as nearly as possible with the same groups in the Census we see, however, that almost all groups remain underrepresented - the comparator proportions in the UK population as a whole being larger.[7] (It remains the case, however that these comparator populations contain both British and non-British nationals.)


Table 2: Distribution of British nationality academic staff, aged 25 to 44 years, by ethnic group. The UK census data (25 to 44 years) is given for comparison
Ethnic GroupAcademic Staff*UK Population**
n%%
White45,58896.394.5
Non-white1,7513.75.5
Chinese2570.50.4
Indian6111.31.8
Pakistani1830.40.8
Bangladeshi360.10.2
Asian (other)2900.60.5
Black African1250.30.5
Black Caribbean / Black (other)2490.51.3
Total47,339100100
* Data source: HESA, 1996

** Data source: UK Census, 1991


9.4
The fact that non-British non-whites (and non-British whites) are a distinct employment group is shown in Table 3 in which the concentration of non- British staff in 'research-only' staff is clear.


Table 3: Primary employment function of academic staff, by ethnicity / nationality group
Ethnicity / Nationality GroupTeaching OnlyResearch OnlyTeaching and ResearchAll Job Types
n%n%n%n%
White, British9,29111.219,20323.254,41765.682,911100
White, Non-British6767.34,19245.14,41947.69,287100
Non-White, British2138.888636.61,31954.62,418100
Non-White, Non-British1394.81,65256.61,12638.62,917100
All Staff10,31910.625,93326.661,28162.897,533100

9.5
Many of these are overseas-origin postgraduate and postdoctoral academics who are engaged in time-limited research projects and are typically in Science, Engineering and Medical Schools in pre-1992 universities. The non-white and non-British are therefore a distinct presence in UK universities, attracted to and recruited into research-based career- development posts in the research-rich old universities. In employment function the British non-whites are rather closer in profile to the academic staff as a whole but they too are over-represented in Research-only posts, being 37 per cent Research-only employees compared to 27 per cent among all staff. Table 4 shows further that ethnic/nationality groups are not evenly distributed through subject areas. This is a broad indication of how market pressures may be operating in the academic marketplace. Among non-white non-British, 31.1 per cent are in Engineering, whilst 'local' staff may be hard to recruit; in fact 6.7 per cent of all Engineering academic staff are non-white non-British, double their representation in academic staff as a whole. By contrast very small percentages of non-whites (both British and non-British) are in Arts subjects; summing British and non-British, Arts subjects are 98 per cent white staff, the highest proportion of any subject area.


Table 4: Subject area of employment of academic staff, by ethnicity / nationality group
Ethnicity / Nationality GroupMedicineScience
EngineeringSocial Science / EducationArtsHealthAgricultureServicesTotal
n%n%n%n%n%n%n%n%n%
White, British10,27812.418,23922.011,21713.522,37927.011,86914.36,6778.11,5551.96970.882,911100
White, Non-British1,41415.22,70829.21,05011.31,97321.21,73818.72662.9971.0410.49,287100
Non-White, British59224.546019.040716.845118.71235.135914.8180.780.32,418100
Non-White, Non-British48416.680027.490631.143014.71254.31495.1190.740.12,917100
All Staff12,76813.122,20722.813,58013.925,23325.913,85514.27,4517.61,6891.77500.897,533100

9.6
There are important differences when we look at individual subject areas. The difference between minority presence in the highest represented subjects - 14 per cent in Chemical Engineering - is much greater than in the lowest - 1 per cent in Geography.

Career Experience of Minorities in British Academia.

10.1
Recruitment into the profession reflects prior social experience. The profession now mostly demands doctorates as a requisite, and educational progression and cultural capital generally, are influenced by class points of origin. To this extent representation in the profession is a different kind of measure from career experience within it. In the HESA data there are two principal measures which enable us to gauge the latter: the type of contract on which staff are employed and seniority.

Fixed Term Contracts.

11.1
The employment of staff on fixed term contracts has increased within UK universities and colleges in recent years (Times Higher Education Supplement August 29 1999). Permanent staff have a better base from which to plan their career development, these posts give greater security and are regarded as more desirable. However, although career development for contract researchers remains a contentious issue, it is certainly true that in research-only posts fixed term contracts are the norm. In the HESA data for 1996, for those with information on ethnicity and nationality (N=97,533), 93 per cent of Research Only posts were fixed term contracts and they are overwhelmingly concentrated in the pre-1992 universities.

11.2
Table 5 below compares percentages of staff on permanent contract, by ethnic/nationality group, for all types of employment. Minority British are more likely than whites of British nationality to have fixed term contracts; non-British minority staff are more likely to have fixed term contracts than non-British whites.


Table 5: Distribution of academic staff by ethnicity / nationality group and contract type
Ethnicity / Nationality Permanent contractsFixed-term ContractsOther Contracts*All Staff
Groupn%n%n%n%
White, British53,41364.427,93533.71,5631.982,911100
White, Non-British3,50837.85,67661.11031.19,287100
Non-White, British1,22650.71,16448.1281.22,418100
Non-White, Non-British93232.01,97267.6130.42,917100
Total59,07960.636,74737.71,7071.797,533100
* These include casual and other contracts

11.3
The difference between the British (white and non- white) and non-British (white and non-white) is largely attributable to the concentration of non-British staff in temporary contract 'Research-only' posts (non-British whites 45 per cent, non-British non-whites 56 per cent) and over 95 per cent of Research-only posts are fixed term contracts. But what explains the 13 per cent difference between white and non-white British academic staff in permanent contracts? As we have suggested, employment type (research only, teaching only, teaching and research) is critical and in fact explains 85 per cent of the difference in contract type. Teaching and Research posts are the ones where permanent contracts are commonest; Table 6 below looks at differences between ethnic/nationality groups in contract type, looking within Teaching and Research posts.


Table 6: Distribution of academic staff in teaching and research posts only, by ethnicity / nationality group and contract type
Ethnicity / Nationality Permanent contractsFixed-term ContractsOther Contracts*All Staff
Groupn%n%n%n%
White, British45,78884.28,29115.23380.654,417100
White, Non-British3,14771.21,25728.5150.34,419100
Non-White, British1,04279.027420.830.21,319100
Non-White, Non-British80071.032428.820.21,126100
Total50,77782.810,14616.63580.661,281100
* These include casual and other contracts

11.4
It is clear that controlling for type of employment narrows the gap in permanence of contract for white and non-white British, reducing the difference from 14 per cent to 5 per cent. Part of the story of ethnic disadvantage in higher education must be about why minorities get into 'research only' or 'teaching only' posts where fixed term contracts are commoner, as well as raising the question of why, even in 'teaching and research' posts, minorities are more likely to get fixed term contracts.

Sectors, Gender and Ethnicity, Age and Subject Groups: Four Dimensions of Inequality and Ethnicity

12.1
Despite this apparent partial 'explanation' of the contract position of non-white British staff, important inequalities reappear when we inspect four things: sectors of higher education, gender and ethnicity, age and subject areas.

Sectors.

13.1
An inspection of differences between sectors reveals that the pattern in table 6 above is composed of a 'beneficial' position for British non-whites in post-1992 universities and in other colleges, and a disadvantaged position for British non-whites in pre-1992 universities and medical schools. Whereas in the new universities British non-whites in teaching and research posts are rather more likely to be in permanent posts than British whites, in old universities and medical schools the minority disadvantage reappears; the same pattern obtains for non-British staff. In this table we show old and new universities only, where the great majority of staff are employed.


Table 7: Distribution of academic staff in teaching and research posts only, by ethnicity / nationality group, contract type and university sector
Old UniversitiesNew Universities
Ethnicity / Nationality Permanent contractsFixed-term ContractsPermanent contractsFixed-term Contracts
Groupn%n%n%n%
White, British24,68582.35,27117.616,15488.91,92010.6
White, Non-British2,15668.698031.278281.717017.8
Non-White, British38968.218131.854590.7549.0
Non-White, Non-British42764.423635.631885.75214.0
Total27,65780.56,66819.417,79988.62,19610.9
Note: Staff on non-permanent, non-fixed term contracts (n=358, no greater than 0.5 per cent of any of the ethnicity/nationality groups) have been omitted from this table. Percentages across rows therefore do not quite add up to 100 per cent.

Gender and Ethnicity

14.1
Women in all ethnic categories are more likely than men to have fixed term contracts White women are less well placed in contractual terms than white men; non-white women are less well placed than either white women or non-white men. Non-whites are, overall, more likely than whites to be in fixed term contract posts. These two patterns of disadvantage - gender and ethnic/national - require considerable unravelling. Table 5 above shows the simple overall position of ethnic disadvantage; the overall gender disadvantage is shown by 48 per cent of women in permanent contracts compared to 61 per cent of men.

14.2
The non-British, white and non-white, are, as we observed above, much more likely to be on fixed contracts. British non-whites are also less likely than British whites to have permanent contracts and the difference between them (13 per cent) is the same as that between men and women. It is, however, important to look at age as a factor in permanence of contracts since we know that younger staff and recent entrants to the profession are more likely to have fixed term contracts, and we know that the older the academic staff, the more white and male it becomes. Table 8 below shows the percentage of staff on permanent contracts, by age groups and by i. ethnicity/nationality and ii. gender.


Table 8: Proportion of staff on permanent and fixed-term contracts by age group, ethnicity / nationality group and gender
White, BritishNon-White, BritishWhite, Non-BritishNon-White, Non-British
Permanent contractsFixed-term contractsPermanent contractsFixed-term contractsPermanent contractsFixed-term contractsPermanent contractsFixed-term contracts
n%n%n%n%n%n%n%n%
Age Group
18-302,51817.811,36780.57311.953587.329111.22,28587.84913.032686.2
31-4012,77157.78,98140.642147.745051.01,36634.42,56764.742824.01,34675.6
41-5021,18581.34,45417.146578.412220.61,07964.058134.531154.525845.2
51-6014,85084.82,33013.323184.04215.366678.417220.213078.33521.1
61+2,07269.576325.63672.01224.010564.05634.11470.0630.0
All Ages53,39664.527,89533.71,22650.81,16148.13,50737.85,66161.193232.01,97167.6
MaleFemale
Permanent contractsFixed-term contractsPermanent contractsFixed-term contracts
n%n%n%n%
Age Group
18-302,07413.912,63784.51,52715.28,34383.0
31-4011,75648.911,87849.46,52246.77,02450.3
41-5019,67483.13,62615.38,02567.53,48729.3
51-6015,82586.92,06711.33,43770.01,21925.1
61+2,49068.595126.230560.614328.4
All Ages51,81961.331,15936.919,81648.020,21649.0
Note that the data shown by gender include all staff, even those who did not record their ethnicity / nationality. We have excluded 135 staff from this table with unknown age from the ethnicity analysis and 360 staff with unknown age from the gender analysis. Casual and other contracts are excluded from the tables since they are very small in number; thus adding permanent and fixed contracts in most cases yields about 97%, rather than 100%.

14.3
The pattern of disadvantage by contract type is different for women in relation to men, as against non-white British in relation to white British. There is clear overall gender and ethnic/national inequality in contracts: both minority groups (women and non- whites in Table 8) are more likely to have fixed term contracts. But for ethnic minorities this is evident among the young recent entrants to the profession (18-40 years old). In the older age groups the non-white disadvantage is not evident. Among females the contract disadvantage is virtually nil in the young age groups (18-40 years old) but is marked in the middle-older age groups i.e. the almost exact opposite to the ethnic minority pattern.

14.4
We also know that fixed term contracts are prevalent in research only posts and least common in Teaching and Research posts. Table 9 removes the possible effect of employment function by looking only at Teaching and Research posts. The pattern remains basically the same, that is disadvantage for non-white British among young staff which is erased as we move up the age groups; and nil difference for females in the youngest academic staff, but evident among the older staff. The main difference is that for teaching and research posts the disadvantage for women 'appears earlier'; the pattern of 'no disadvantage for young females' but with disadvantage re- appearing with age, is marked at 31-40 rather than 41-50 years old.


Table 9: Proportion of staff in teaching and research posts on permanent contracts, by age group, ethnicity / nationality group and gender
White, BritishNon-White, BritishWhite, Non-BritishNon-White, Non-British
Permanent contractsFixed-term contractsPermanent contractsFixed-term contractsPermanent contractsFixed-term contractsPermanent contractsFixed-term contracts
n%n%n%n%n%n%n%n%
Age Group
18-301,95255.01,54843.65543.37256.723945.328654.24047.14350.6
31-4010,83078.02,97421.436173.412926.21,21464.665935.136762.721837.3
41-5018,07489.52,02410.039690.0439.898583.319416.426483.05417.0
51-6013,04290.71,2498.719991.3198.761387.78512.211794.475.6
61+1,87578.448520.33175.61024.49575.43023.81285.7214.3
All Ages45,77384.18,28015.21,04279.127320.73,14671.31,25428.480071.032428.8
MaleFemale
Permanent contractsFixed-term contractsPermanent contractsFixed-term contracts
n%n%n%n%
Age Group
18-301,62451.41,49747.41,19751.81,07946.7
31-4010,34275.73,25123.85,37971.02,12128.0
41-5017,36091.41,5588.26,64182.01,38417.1
51-6014,27591.81,2007.72,87083.951014.9
61+2,28377.462721.326478.66619.6
All Ages45,88484.58,13315.016,35175.15,16023.7
Note that the data shown by gender include all staff, even those who did not record their ethnicity / nationality. We have excluded 34 staff from this table with unknown age from the ethnicity analysis and 62 staff with unknown age from the gender analysis.

14.5
We can conclude several things from this:

  1. contract disadvantage for non-white British and for females represents two divergent patterns when we control for age.
  2. ethnic disadvantage is not explained by youth or recency of entry to the profession; young non-whites (British) are contract-disadvantaged compared to whites.
  3. by contrast young females appear to be entering the profession on similar terms to males; their disadvantage overall is composed of disadvantage largely in the middle and older age groups.
  4. ethnic disadvantage and gender disadvantage are not explained by employment type; the disadvantage remains evident within the posts least likely to be on fixed terms.

Subject Areas.

15.1
In Table 4 above we examined subject area of employment of academic staff by ethnicity/nationality groupings. We are now interested in whether subject area has any influence on differences in type of contract of employment. Looking back to Table 4 we see that non-whites are very considerably concentrated in three subject groups: compared to all staff of whom 27 per cent are in medical and engineering subjects, 41 per cent of non-white British and 48 per cent of non-white non-British are in these two subject areas, Of non-white British a further 15 per cent (compared to 8 per cent of all) are in Health subjects. This offers a further clue to the contract-disadvantage of non-whites since engineering and medical subjects (but not Health) are areas where fixed term contracts are commonest, particularly among young Research-only staff where we know non-white representations are higher. For this reason we must look at contract differences across subject areas, between whites and non-whites controlling for age and employment function. Table 10 below shows the percentages of staff on permanent contracts by ethnic/nationality groups and subject area, within the younger staff and Teaching and Research posts only.


Table 10: Proportion of staff in permanent contracts (aged under 45 years in teaching and research posts only) from each ethnicity / nationality group in major subject areas
Ethnicity / Nationality GroupMedicineScience
EngineeringSocial Science / EducationArtsHealth
n%n%n%n%n%n%
White, British1,48851.93,78681.72,49085.55,47878.82,96178.52,24182.2
White, Non-British13436.137273.117874.259065.250064.97977.5
Non-White, British5535.39579.812589.915476.63965.010285.7
Non-White, Non-British3726.69966.915276.815772.03151.75288.1

15.2
With respect to non-white British Table 10 suggests that the pattern of contract disadvantage is quite different in the three key areas - medical, engineering and health groups of subject - where more than half of non-whites are to be found. Table 10 shows only younger staff and staff in Teaching and Research since age and employment type influence contract type very considerably. In Health subjects non-white British are rather better placed than white British; in Engineering subjects the non-white British are better placed than white British but both non-British groups of staff are less likely to have permanent contracts; but in Medical subjects a considerably higher proportion of non-white British staff are in fixed term contract posts. With the exception of Health subjects, non-white non-British are less well-placed than all other groups. It is also notable that in Arts subjects both groups of non-whites (and non-British whites) are more likely to have fixed term contracts than British whites. This indicates that part of the 'contract disadvantage' of ethnic minority staff is linked to representation in subject areas where fixed contracts are more common, particularly in Medicine.

Seniority

16.1
Seniority is the second indicator of position within the profession available to us from the HESA data set. Professors and Senior and Principal Lecturers are classed as high grade posts, lecturers, researchers and others classed as low grade.[8] The different grade systems of the pre- and post-1992 universities pose a problem and so we will (see below) examine ethnic/national distributions by grade within those two sectors. But first we show the grade distribution by ethnic/national groups controlling for age for all academic staff.


Table 11: Proportion of staff in promoted posts (senior lecturer / professor) from each ethnicity / nationality group in age groups
Ethnicity / Nationality Group31-4041-5051-6061+All Ages
% in promoted posts% in un-promoted postsTotal count% in promoted posts% in un-promoted postsTotal count% in promoted posts% in un-promoted postsTotal count% in promoted posts% in un-promoted postsTotal count% in promoted posts% in un-promoted postsTotal count
White, British13.786.322,13937.462.626,04550.249.817,50954.445.62,98128.072.068,674
White, Non-British8.092.03,96832.467.61,68548.052.085051.248.816414.785.36,667
Non-White, British11.188.988226.373.759341.158.927548.052.05016.383.71,800
Non-White, Non-British4.395.71,78117.382.757134.365.716645.055.0208.391.72,538
Note: The youngest age group has been excluded, since so few of these staff are in promoted posts. A further 133 staff with no recorded age were also excluded.

16.2
In all age groups fewer Non-white British staff have been promoted than white British; and, among the non-British, the white staff also do markedly better than non-whites. Indeed with one small exception (White non-British at ages 31-40) the percentage of staff promoted in each category presents a perfect descending amount from white and British to non-white and non-British.

Seniority, Sectors and Type of Post

17.1
There are good reasons for examining differences in seniority within sectors and within Teaching and Research posts only. Sectors differ in the way they classify senior posts and the 'Professor plus Senior Lecturer or Principal Lecturer' (classed as high-grade posts) category captures a considerably lower proportion of staff in New universities (21 per cent) than in the old universities (34 per cent). Again it is more prudent to compare ethnic differences in seniority within the 'teaching and research' posts, to be sure that we are comparing like with like.


Table 12: Proportion of staff in promoted posts (senior lecturer / professor) in age groups by ethnicity / nationality group, university type (old vs. new) and primary employment function (all vs. teaching and research only)
Ethnicity / Nationality 31- 4041-5051-6061+
Groupn%n%n%n%
ALL POSTS
Old Universities
White, British1,97015.36,26750.06,35464.81,33769
White, Non-British2618.542638.633056.57462.2
Non-White, British5912.69238.07057.31768.0
Non-White, Non-British483.56318.64447.3861.6
New Universities
White, British73511.72,49326.81,70631.918627.1
White, Non-British285.17719.04825.3413.0
Non-White, British2810.05119.03226.0527.8
Non-White, Non-British217.13017.31017.5120.0
TEACHING AND RESEARCH POSTS
Old Universities
White, British1,79223.75,95058.66,13370.81,25674.2
White, Non-British22516.339551.131763.66973.4
Non-White, British5022.08050.66769.81777.3
Non-White, Non-British4211.36037.94363.3770
New Universities
White, British61413.22,13128.61,48134.516632.5
White, Non-British246.45521.44227.6312.0
Non-White, British2310.74721.02625.0426.7
Non-White, Non-British169.22317.91022.2150.0
Note: Subjects with missing data on age are omitted from this table (For old universities, all posts, n=41; new universities, all posts, n=69; old universities, teaching and research posts, n=16; new universities, teaching and research posts, n=13.

17.2
The evidence of Table 12 suggests that when we look within the two largest sectors (pre and post-1992 universities) the pattern of disadvantage in seniority for non-whites is sustained. The exception, in comparing all posts, is in the oldest age category (61 and above) in which the numbers of non-white British staff are very small. In comparing seniority by ethnic groups in the Teaching and Research posts only we are looking at the employment type with the most stable career pattern - i.e. where staff are least likely to have fixed term contracts and are most likely to stay in posts long enough to get promoted. Again the pattern of disadvantage is sustained through the age groups and in both old and new universities, with the exception of staff 51-60 years old in pre-1992 universities.

17.3
The pattern of ethnic disadvantage in seniority is not readily explained by differences in age profile of white and non-white British staff, by any difference of pattern between sectors, or by type of employment function.

Reported Discrimination and Views of the Prevalence of Discrimination

18.1
Over 400 ethnic minority staff responded to questionnaires delivered to staff in selected UK universities along with a small control sample of white staff (Carter, Fenton and Modood 1999). The minority staff were asked whether they personally had experienced racial discrimination in job applications and in promotion procedures and whether they had experienced racial harassment from staff or students. They were also asked whether they thought discrimination was common in higher education institutions. About one quarter of minority respondents reported that they had experienced discrimination in job applications; this was more common (30 per cent) among non-British minorities. Fifteen per cent said the same about promotions, and nearly one in five reported experience of racial harassment, women being more likely than men to say they had experienced racial harassment.

18.2
It is very difficult to judge how this compares with individual reported experience of discrimination on other spheres. In the Fourth National Survey (Modood et al, 1997) there are no directly comparable questions although several cover the same topics. With respect to experience of discrimination in job applications (in the Fourth National Survey) 19 per cent of all minorities said they had been refused a job for 'racial/religious reasons' and this figure varied from 28 per cent of Black Caribbean respondents to 5 per cent of Pakistani respondents. Some of the respondents to our academic staff survey clearly believed that discrimination in higher education was widespread. Among British minority respondents 55 per cent believed 'greatly' or 'partly' that there is discrimination in employment in higher education and even among whites 19 per cent held this view and a further 41 per cent were 'not sure'. Among all respondents between 35 per cent and 40 per cent were 'not sure' about their institution's commitment to equal opportunities, and 16 per cent of minority academics characterised their institution's commitment to equal opportunities as 'not very' or 'not at all'. Academics from post-1992 universities were more likely to report experience of discrimination and harassment and this was true for men and women. This is consistent with the finding (above) that in the post-1992 sector there are fewer opportunities for promotion, and may indicate that there is greater work pressure and frustration allied to poorer resources and higher teaching loads; and that ethnic minorities are less likely to be promoted compared to their white peers than in the pre-1992 sector.

18.3
In the discussion groups with ethnic minority staff and, separately, with research students, discussants expressed resentment at being typecast by ethnicity, nationality and gender and argued that white academics should reflect on the assumptions they held about minorities. For, they argued, such assumptions and the practices which reflected these assumptions, resulted in the marginalisation of minorities and their concerns. While some discussants talked of their experiences (or experiences of people they knew) of being denied opportunities, especially promotions and managerial responsibility, most of the discussion of discrimination was about stereotypes and subtle assumptions about inferiority or suitability. There was also a strong sense that networks and sponsorship were critical to academic careers.

Equal Opportunities.

19.1
In our survey of institutions, each institution was asked whether they had an equal opportunities policy and what elements (e.g. ethnic monitoring, policies on harassment) these policies contained. This enabled us to class institutions by policy from least developed to advanced policy (Carter, Fenton and Modood 1999) . This shows that the universities and colleges with advanced policies were not all in the regions or sectors that might be expected (e.g., new universities) and it shows that institutions with high proportions of ethnic minorities include some with and some without developed policies. We were also able to 'superimpose' the institutional policy data on the HESA data set[9] and examine the differences between those with the least and most advanced policies. It is entirely possible that the 'measured' (item by item) strength of equal opportunities policies is less important than the less tangible cultural ambience of an institution, irrespective of its formal policy. Certainly the analysis of the position of non-white staff by the 'policy-type' of their institutional employer, does not show any difference in the contractual terms of those staff. Table 13 below shows that among the approximately two thirds of non-white staff who were 'covered' by our institutional data, those in institutions with developed policies were only just better placed with respect to contracts compared to those in institutions with less developed policies. This table collapses more and less developed policies into two categories.


Table 13: Proportion of non-white staff with permanent / fixed-term contracts, by equal opportunity policy development
Policy of Institutional EmployerPermanent contractsFixed-term contracts
n%n%
Less Developed46335.384664.5
More Developed76836.91,29762.3
Note: Staff on non-permanent, non-fixed term contracts have been omitted from this table. Percentages across rows therefore do not add up to 100%

Conclusions

Closure and Gatekeeper Discrimination.

20.1
Our earlier brief review suggested that some 'democratisation' of higher education in Britain has been taking place, particularly since the 1960s. Nonetheless as social closure begins to lose its strength, certainly by contrast with the 'categorical' exclusion by religious group in the nineteenth century, we may hypothesise that strains of culturally defined suitability and acceptability criteria almost certainly persist, with varying strength in different disciplines and institutions. An overt gatekeeper discrimination of a "they shall not pass" kind becomes unsustainable as a public meritocratic ethos is adopted; but tacit and less detectable discriminations occur. The pattern is certainly not one of total closure or exclusion. In the younger compared to older aged staff the proportion of ethnic minority staff is higher, and in some areas and among some groups, recent entry to the profession is higher than one would predict from population proportions. (See Table 15 below on ethnic group differences).

20.2
This is not to say that discrimination on racial grounds does not take place; a considerable proportion of academic staff, both white and non-white believe that it does (Carter, Fenton and Modood, 1999). But it does suggest, as our limited qualitative enquiries indicate, that it more often takes the form of tacit assumptions of a cultural and acceptability kind and reflects experience within the profession rather than entry. The national academic staff records cannot prove discrimination, only 'unexplained' patterns of inequality. But the very fact that ethnic minorities enter areas where white applicants have declined (see below on markets and Table 4 above) suggests that, conversely, where white demand is stronger, minorities find it harder to enter.

Equal Opportunities

20.3
The 'philosophy' of equal opportunities has two distinctive features. First it is the public and political declaration of the paramountcy of the merit principle; it is needed wherever patterns of culturally defined exclusion are found. Secondly it is a declaration that meritocracy does not operate by default. It is a declaration that, for meritocracy to work, the steps which 'guarantee' it have to be built minutely into the selection and promotion procedures of the institutions. Although a pattern of discrimination in the broadest sense may be the rationale for the implementation of equal opportunities procedures, this need not be read as discrimination in the 'popular' sense of the 'prejudiced gatekeeper'. A clear example in the field of gender and equal opportunities would be the need to take account of family obligations in the judgement of women's careers; failure to do this would discriminate against some women, but not in the simple sense of the 'prejudiced gatekeeper'. Recognising the complexity in discrimination enables us at least in part to explain the common disjuncture between the belief that discrimination is widespread and the denial that it occurs.

20.4
There is no compelling evidence in this study that equal opportunities policies may create more meritocratic institutions. Formal policies are unlikely to be decisive. For example the pattern of 'sponsored mobility', so characteristic of academic life (Heward et al.1997) is unlikely to change quickly. Academics are drawn towards 'sponsorship' judgements not just because they prefer 'acceptable' colleagues (however they might define that) but also because academic judgements are so difficult to make. In circumstances of uncertainty people may be especially dependent upon the judgement, as offered in confidential formal references and unwritten, perhaps even 'unsaid' remarks, of senior colleagues. If 'outsiders' need equal opportunities policies and sponsors, then the small numbers of non-whites in UK academic institutions may make for a shortage of sponsors for some years .

Institutional Racism

20.5
The concept of institutional racism is the most wide reaching and all-embracing, this being both its strength and its weakness. Institutional racism has in its career as a concept incorporated three ideas: that the culture of an institution expresses both overt and tacit racialised ideas and attitudes, and that this cultural frame goes largely unchecked; that the routine practices of an institution, by intention and by inertia, act to the disadvantage of excluded or partially excluded groups; and thirdly that the structured disadvantage of groups in society at large is reproduced in the institution, so long as it takes no account of them. As a diagnosis which implies remedies, institutional racism is so wide in its embrace that an equally wide range of cures are called for. The first dimension suggests the need to alter the culture of an institution; before this can happen, the culture must be understood. The second dimension implies the need for self-conscious implementation of equal opportunities and the questioning and revision of routine practices. The third dimension implies the need for a society-wide reduction of inequalities and programmes of positive action by institutions. The institutional racism model thus overlaps with an equal opportunities model which demands self-conscious meritocracy in spirit and in procedures. It also overlaps with the market disadvantage model by pointing to the 'generalised' disadvantage of excluded or partly excluded groups. Institutional racism however stands quite alone in pointing, in a totalising way, to the cultural environment of an institution. This is its greatest significance for universities.

20.6
The research reported here did not incorporate a study of the culture of institutions tout court - a task quite beyond our means. But the study groups with staff gave a glimpse of day-to-day cultural events: a scientist whom we interviewed told us how his colleagues characterised his 'Indian-ness' as a difference which misrepresented his own identity, and a Chinese respondent talked of a kind of exclusion from the social life of her department. Scientists told us how they might be valued as laboratory team members, but expected - and experienced - presumptions against appointing them as team leaders. Furthermore universities have never explored their own postcolonial nature: how far have universities emancipated themselves from a colonial hierarchy of valued institutions and qualifications? This dimension of university life is only hinted at in our research; future research will almost certainly show its importance.

Markets: Disadvantage and Choice

20.7
The market disadvantage model suggests that excluded groups are likely to be generally disadvantaged rather than specifically disadvantaged in the institutional arena in question - in our case, academic employment. Individuals from specific groups will be all the less likely to seek work in highly credentialised occupations if they have already experienced prior disadvantage in acquiring the credentials. Social mobility occurs in both white and non-white populations but, equally, class advantage in all ethnic groups exerts an influence on the acquisition of social and cultural capital and on the deployment of tactics for entry into favoured employment and business. Any earlier evidence of the acquisition of educational capital will provide some measure of advantage and disadvantage and entry to university is one such indicator. The data for university admissions in 1992 is the earliest available to us with the necessary detail. Though not ideal it is a valuable pointer to a pattern of educational advantage and the acquisition of credentials. Table 14 clearly shows that the pattern of advantage in home admissions to university by ethnic groups in Britain (1992)[10] almost perfectly matches the 1996 pattern of representation in the academic profession of staff under 45 years.


Table 14: Proportion of UCAS admissions and academic staff, ranked by ethnic origin
EthnicityUCAS admissions*Academic Staff**
Asian (other)+1611.20
Chinese+1091.25
Indian+190.72
Black African+140.60
Pakistani- 190.50
Bangladeshi- 450.50
Black Caribbean / Black (other)-630.38
* UCAS admissions in relation to population representation, see Modood et al. 1994
** Staff ethnicity distribution as a proportion of UK population aged <45 years. Calculated from figures in Table 2

20.8
The market choice model by contrast highlights the choices made by privileged groups, who may abandon a previously closed arena if it loses its attractions; and the choices of 'disadvantaged' actors, who may acquire credentials - human capital - and invest it in the arenas which are more open, more rewarding or both.

20.9
The evidence supporting this model is persuasive. First of all it is clearly valuable to regard the academic market as a labour market like any other, that is as one commanded by scarcity, demand and rewards (Guardian, February 24th, 2000). We know that very few postgraduates in leading economics research programmes are UK origin; to fill these places universities have recruited research students from a global base. In many areas of academic life - accounting, medicine, law and engineering - the corresponding commercial field of practice is much more lucrative and makes recruitment difficult (cf. IRS 2000). We have seen in this study that non-whites are better represented in medical posts than other areas; and that the field with the highest non-white staff is Chemical Engineering, another field with high commercial drawing power.

20.10
At the same time the ethnic minority actors themselves are making choices. Medicine itself is clearly valued as a profession among Indians even if the academic teaching path is less sought after by whites. Ethnic minorities are roughly twice as likely to enter Higher Education as whites and on the whole are even more likely to study vocational subjects and those subjects which are of a premium outside academia. Further, there is evidence that some minorities are entering professions other than academia and are beginning to be quite strongly represented (Esmail and Everington 1993; MacKenzie 1995; Shiner 1997). For example, and significantly, in relation to our findings about medicine in higher education, 30 per cent of NHS hospital doctors in 1998 were not white, nearly half being Indians, although to compare directly with our analysis of Higher Education we need to note that many non- white National Health Service doctors are from overseas. Or, to take law: 5.4 per cent of solicitors in private practice, and 16 per cent of new trainees, were ethnic minorities; amongst barristers, the corresponding figures in 1999 were 8.8 per cent and 17 per cent, and in each case more than half of these percentages were accounted for by South Asians. If academic careers are less attractive, it is unlikely to do with perceptions of relative discrimination since all major institutional areas, including medicine and law, are seen to discriminate. It is probably because minorities with educational capital are investing this capital in better rewarded arenas; if so they would be making the same market choices as whites in a period when the attractiveness of academic life as a profession declines steeply (Guardian February 25th, 2000 'Universities research brain drain'). The differences between specific ethnic groups in the areas of academic life also reflect preferences and 'traditions' of employment. Table 16 below shows that the distribution of any ethnic group through the main subject areas is very uneven. The representation of Indians in medicine we have already discussed; the numbers of Black Caribbean staff in Health studies may also reflect historical employment in the National Health Service.


Table 15: Ethnicity of academic staff by subject area of employment
EthnicityMedicineScience
EngineeringSocial Science / EducationArtsHealthAgricultureServicesAll
n%n%n%n%n%n%n%n%n%
White11,69212.720,94722.712,26713.324,35226.413,60714.86,9437.51,6521.87380.892,198100
Black Caribbean 3511.3278.7278.78427.1216.811436.810.310.3310100
Black African10018.09316.711120.014025.2285.07713.861.110.2556100
Black (other)1810.82112.6148.44627.52012.04728.10010.6167100
Indian41329.730021.625218.127619.9523.7785.6141.050.41390100
Pakistani8830.26723.05318.25518.9124.1103.462.100291100
Bangladeshi1824.02026.72026.71013.345.322.711.30075100
Chinese26816.051730.961536.81448.6442.6764.560.430.21,673100
Asian (other)13615.621524.622125.312614.4677.710411.930.310.1873100

20.11
There are five groups - Indian, Pakistani, Bangladeshi, Chinese and Asian other - where around 70 per cent of the academic staff in that group are in three broad areas, Medical, Science and Engineering subjects. This compares with 48 per cent of white staff. By contrast only 28 per cent of Black Caribbean staff are in the Medical, Science and Engineering group, whilst 64 per cent are in two areas - Sociology and Education, and Health studies.

20.12
Finally, we may look at the models of ethnic advantage and disadvantage discretely and judge their explanatory power in the field of academic staff. Such a conclusion might look like that represented in the chart below:


Table 16: Models of ethnic disadvantage in academic staff employment
Social ClosureHistorically powerful, in decline
Gatekeeper DiscriminationA factor, more likely by exercise of tacit assumptions
Institutional RacismOverlaps with all other models; on its own is most powerful as a model of institutional culture exercising a pervasive effect, especially in particular disciplines
Equal OpportunitiesFormal policies may have marginal influence; unlikely to be decisive without allied factors
Market Disadvantage and Market ChoiceMost powerful explanatory model, in tandem with institutional racism qua a model of institutional culture

20.13
However the most satisfying model comes from a combination of elements of two models: institutional racism and a market model. The account of higher education and the academic profession will read like this:

The institution has a history of declining social closure but continues to have a culture of learning and scholarship which is informed by a hierarchy of places and styles of learning which are often informed by ideas of social suitability or acceptability. The workplace culture and the unreflected routine practices of academic life combine to exclude some and retard the progress of others. But the will to guard the gates of entry and promotion are weakened by the declining attractiveness of the profession and the need to broaden recruitment in a massified system. New groups are differentially placed with respect to educational capital to take advantage of this market opportunity and even those who are best placed may choose to invest their human capital elsewhere. But, for some, the new market situation provide opportunities which they are well placed to take.

20.14
The pattern of disadvantage described here has similarities with class and cultural inequalities characteristic of higher education past and present. Some of the disadvantages of social class are repeating themselves in combination with disadvantaged ethnicities. If the class culture of the academic profession has been partly undermined by 'democratisation', the cultural and political climate with respect to ethnic diversity has only just begun to be considered.

Notes

1Funding for the project came from Higher Education Funding Councils for Scotland and for England, Commission for Racial Equality, The Association of University Teachers, The Committee of Vice Chancellors and Principals, and Natfhe, the University and College Lecturers' Union

2One was a survey of institutions, asking for information about their Equal Opportunities Policies. The second was a questionnaire survey of academic staff.

3The Higher Education Statistics Agency (HESA) data are based on records collected from Higher Education institutions annually. All the analysis in this paper are based on the 1996-97 data set although later the 1997-98 set was made available to us. We are grateful to the Higher Education Funding Council for England for advice on the data set. HEFCE and HESA cannot be held responsible for the analysis in this paper. The HESA data set total with duplicates removed is 126,142 and the total for whom ethnicity and nationality is known is 97,533, the latter figure being the one on which most analysis in this paper is based. 'Subject area' was identified by the academic cost centre 'Primary employment function' was identified by the field of that name which is classified 'Teaching only', 'Research Only' and 'Teaching and Research'. 'Seniority' (of grade) was identified by the field grade structure from which 30 grades are classed into 'Professor', 'Senior Lecturer and Researchers', 'Lecturers', 'Researchers', and 'Other Grades' the first two of which are classed as high grade, referred to in the text as promoted or senior grades. Contract (permanent, fixed term, and casual and other) was identified by terms of employment. Age reference date was 1st August 1996.

4If substantial numbers of these unrecorded for ethnicity were 'non-whites' the conclusions about representation would be altered. But there is no reliable way of estimating the likely ethnicity of staff for whom ethnicity is 'other and unknown'. Those with 'ethnicity other and unknown' are roughly equally likely to be male or female but more likely to be in old universities, on fixed contracts, to be younger and to be non-British nationality. We have proceeded on the basis of the records of the 77 per cent of HESA-recorded staff for whom ethnicity and nationality is known.

5Imperfect because some long-term settled non-whites may not have British nationality; and because some not having British nationality may in effect be settled in the UK.

6Length of service and age are very closely correlated. Both have deficiencies as variables. An older person may have only just entered the profession. And duration of service in post only relates to the staff's current employ. Thus a professor recently appointed at university would be recorded as a short duration employee. On balance age is a better variable to use as a control in measuring advantage in the profession.

7For tables 1 and 2, with such a large data set all differences are statistically significant at the 0.01 level. More important is to inspect the size of the differences. Confidence intervals are not appropriate since both data sets are 'complete populations' not samples. In all subsequent tables from table three onwards, all chi squared results are significant at the .01 level.

8Possible differences in the recording of staff as Professors has meant that individual universities have contested some published accounts of seniority and ethnicity.

9126 institutions had responded; the classification of policy development for the institution was attached to the individual record of the employed staff, thus giving us institutional data for about two thirds of the HESA data set.

10 The Modood and Shiner (1994) 'ranking' of educational attainment was based on entry to university places as a proportion of 18 year olds.

Acknowledgements

The authors would like to thank Mona Okasha (Department of Social Medicine, University of Bristol) for her patient and much-appreciated work on the tables in this paper. We would also like to thank Pete Blair and Linda Hunt (Department of Child Health, University of Bristol) for statistical advice.

References

BALLARD R, ed., 1994, Desh Pardesh: the South Asian presence in Britain, London, Hurst.

BANTON M., 1983, Racial and Ethnic Competition, Cambridge, Cambridge University Press.

BONACICH E., 1976, Advanced Capitalism and black/white relations in the United States: a split labor market interpretation, American Sociological Review, 81,: 601-28.

HAMILTON C.Vand CARMICHAEL, S. 1969 Black Power: the politics of liberation in America, Middlesex, Pelican Books.

CARTER J. Fenton S. and Modood T., 1999, Ethnicity and Employment in Higher Education, Policy Studies Institute, London.

COHEN, P. (1995) ed., For a multicultural University: working paper 3, London, New Ethnicities Unit, University of East London.

DAHYA B., 1974, The Nature of Pakistani Ethnicity in Industrial Cities in Britain, in Cohen A. ed., Urban Ethnicity, London, Tavistock.

ESMAIL A and EVERINGTON, S., 1993, Racial discrimination against doctors from ethnic minorities, British Medical Journal, vol. 306, 13 March pp. 691-692.

FENTON S., 1996, Counting Ethnicity: social groups and official categories, in Levitas R. and Guy W. (Eds.) Interpreting Official Statistics, London and New York, Longman.

Guardian February 25th 2000, Universities research brain drain HALSEY, A.H. (1992) The decline of donnish dominion, Clarendon Press, Oxford.

HEWARD C., Taylor P., and Vickers R., 1997, Gender Race and Career Success in the Academic Profession, Journal of Further and Higher Education, 21, 2.

JENKINS R., (1986) Racism and Recruitment: Managers, Organizations and Equal Opportunities in the Labour Market, Cambridge University Press, Cambridge.

JEWSON N. and MASON D. (1986) 'The theory and practice of equal opportunities: liberal and radical approaches' Sociological Review, vol.34 no. 2.

MACPHERSON of Cluny, Sir William, 1999, The Stephen Lawrence Inquiry, Stationery Office, London.

MACKENZIE K.J., 1995, Racial Discrimination in Medicine, British Medical Journal, vol. 310, 25 February, pp.478-479.

MCCLELLAND, V.A. (1973) English Roman Catholics and higher education, 1830-1903, Oxford University Press, Oxford.

MODOOD T., Berthoud R., Lakey J., Nazroo J., Smith P., Virdee S. and Beishon S. 1997, Ethnic Minorities in Britain, London, Policy Studies Institute.

MODOOD T and ACLAND, T. eds., 1998, Race and Higher Education, London, Policy Studies Institute.

PARKIN, F., "Strategies of Social Closure in Class formation" in Parkin F. (ed.) (1974), The Social Analysis of Class Structure.

SHINER M., 1997, Entry into the legal professions: the law student cohort study, Year 5, London, The Law Society.

SINGH, Gurharpal, (1998), 'Race Equality staffing policies in Higher Education' in Modood and Acland eds.

SOLOMOS J. and BLACK, L. 1996, Racism and Society, London, Macmillan.

STACEY M., 1990, The Sociology of Health and Healing, London, Routledge.

STEWART, W.A.C. (1989) Higher Education in Post-war Britain, Macmillan, Basingstoke

T. WARNES 1996, The age structure and ageing of the ethnic groups, in D. Coleman and J. Salt Eds. Ethnicity in the 1991 Census, volume one.

WILSON W.J., 1980, The Declining Significance of Race, Chicago, University of Chicago Press.

Copyright Sociological Research Online, 2000

34 59