What factors influence lone parents’ plans for and expectations of the future?
Example project report for the module
SO 201 Surveys, Secondary
Analysis and Social Statistics
(Report produced in March 2000)
Length = 5,990# words
Richard Lampard
(Module tutor)
Abstract:
This project report discusses the findings from a secondary analysis of data from a DSS survey of lone parents carried out in 1989. The report focuses on lone parents’ plans for and expectations of the future. A number of different aspects of lone parents’ lives are considered: future couple relationships, future fertility, paid employment and satisfaction with housing. The relevance of age, gender, class and marital history in relation to these different dimensions is assessed. Age is found to be particularly important; for example, in relation to the likelihood of wanting further children. Differences between divorced, separated and single (never-married) lone parents to some extent reflect the different ages of lone parents in these three categories. However, a multivariate analysis showed that the effects on future couple relationship preferences of a range of explanatory factors (including age and whether a lone parent is divorced, separated or single) persist when other factors are controlled for. A gender difference exists in this specific context, but more generally gender and class do not seem particularly important, at least with regard to some dimensions of the lives of lone parents.
# See the attached covering letter for an important note regarding word length.
University of Warwick, Department of Sociology, 2014/15
SO201: SSAASS: Surveys and Statistics (Richard Lampard)
Points to note regarding this example project report
1. When it was written, a logistic regression analysis was included in this example project to act as ‘icing on the cake’ and to give the project more coherence. Given the emphasis on multivariate analysis in this year’s version of the module, the inclusion of a multivariate analysis or analyses of some sort or other is expected, and there should be a greater emphasis on this/these. Note that the multivariate side of the project might well use three-way cross-tabulations as well as a technique or techniques like logistic regression. In this project the emphasis is quite strongly on cross-tabulations/chi-square and the logistic regression; since many projects focus on categorical variables this is quite likely to be the case for your project too. However, you should certainly consider focusing on or including analyses based on other techniques where appropriate.
2. The example report is very much a function of the specific topic and the way in which I approach writing up data analyses (in terms of style, etc.). While good projects will uniformly be technically competent, they will in other ways be quite diverse, e.g. in terms of presentation, detailed structure, use of literature, etc. Trying to closely mimic this example report may thus not be in your best interest; you probably have your own style of essay-writing, which will need adapting to a secondary-analysis based project, but which you should probably stick to (assuming that it’s any good!)
3. Inevitably, with the benefit of hindsight, there are things about the project and report that I would do differently given the time, or if I were starting from scratch. You should therefore critically evaluate this example project report rather than viewing it as an ‘ideal’; if you think there are better (more sociological, professional, articulate, coherent, etc.) ways of doing things, then do your project/report those ways instead! Arguably it might have been better for it to have more strongly resembled published articles using multivariate analysis techniques…
4. My feeling is that in terms of technical competence, presentation, effort, and content, this example project is a high 2:1 (i.e. 68), although the current emphasis on multivariate analysis within the module also means that it is a bit thin in that respect. Arguably it doesn’t merit anything better than this, because it might be viewed as a bit fragmentary and perhaps also as lacking something conceptually. (However, it’s difficult to evaluate the merits of one’s own work!) What is certainly the case is that people have submitted first class projects in the past; where they have perhaps tended to differ positively from this example project is in the coherence and conceptual sophistication of the way in which the author has addressed the project title.
5. Note that the cross-tabulations are in the text-based format generated by earlier versions of SPSS for Windows (as opposed to, say, Versions 19 or 21). Consequently, they more closely resemble what one gets by ‘exporting’ the contents of the output window/viewer in recent versions than they do the (rather more slick) style of cross-tabulation visible in (and printable from) the Versions 19 or 21 output window/viewer.
In retrospect, the output from the logistic regression is not that well presented…
6. Note that changes in the permissible ‘overshoot’ in terms of number of words (for all Sociology Department undergraduate module assessed work) means that the example project is longer than the maximum length of 5,000 + 10% = 5,500 words, and hence somewhat more detailed than you will be expected to manage.
Richard Lampard (This revision of covering note: 18th October 2014)
i
What factors influence lone parents’
plans for and expectations of the future?
Introduction
This project uses the secondary analysis of data from an existing survey (Bradshaw and Millar 1991) to examine some of the factors influencing lone parents’ plans for and expectations of the future. Lone parenthood has increasingly become an important focus of government and academic social research in Britain, reflecting the fact that “the incidence of lone parent households is now higher than at any time in the last two centuries and is high in comparison with most other European countries” (Rowlingson and McKay, 1998: 1). However, while it has become more common in empirical terms, lone parenthood is still problematised by, among others, politicians and the media, and “a good deal of disagreement has also emerged over the effects of lone mothering on children, and of the costs of lone motherhood to the budget of the state” (Silva 1996: 1).
For many people, lone parenthood represents a transitional stage in the life course, and authors frequently stress the importance of examining the dynamics of lone parenthood (e.g. Rowlingson and McKay 1998: 7). This project broadens out from this focus on the dynamics of lone parenthood to a focus on change more generally, and looks at a number of possible forms of change in the future lives of lone parents: repartnering, taking up work, having further children, and moving house. However, it would be a mistake to view lone parents as a homogeneous group; their diversity is frequently commented upon (e.g. Hardey and Crow 1991: 7). Perhaps the most obvious distinction between different forms of lone parenthood relates to the lone parents’ marital histories, since the situations of never-married lone parents and divorced or separated lone parents are often dissimilar. The project therefore focuses on, among other things, this distinction. (The data source used unfortunately did not permit the plans and expectations of widowed lone parents to be examined).
Lone parents and the future
This section reviews some of the available literature on lone parents, and uses it to identify some topics and themes which can be addressed in the analyses within this project. A keyword search was used to identify a number of recent books on lone parenthood, and from references within these it became clear that extensive official statistical material is available regarding lone parent families within Great Britain, especially in the government publication Population Trends (e.g. Haskey, 1993; Haskey, 1994; etc.).
Official statistics show lone parenthood to be a gendered phenomenon: in 1991 only 9% of lone parents were lone fathers (Haskey 1994). The gendering of lone parenthood is also evident in the emphasis of much of the literature on lone mothers (e.g. Silva 1996; Lewis 1997; Kiernan et al. 1998). To a large extent this emphasis reflects a social policy issue: the extent and desirability of the dependency of lone mothers and their children on the state.
In the early 1990s only a minority of lone mothers were in paid work (38%; see Millar in Silva 1996: 106), and the figure was even lower for never-married lone mothers (Lewis 1997: 62). A continued emphasis on the paid employment of lone mothers is evident in the government’s ‘New Deal’ and is mirrored in official statistics on the topic (e.g. Haskey, 1998b, contains a marked emphasis on lone mothers and work). Paid employment is often seen as a way of reducing lone mothers’ dependency on the state and of ensuring that their children are not in poverty; an alternative ‘strategy’ is, of course, living with a (new) partner.
Lone parenthood sometimes ends as a consequence of children ‘growing up’, but many lone parents move in with a (new) partner within a few years of becoming a lone parent. Rowlingson and McKay note that never-married lone parents cohabit or marry more quickly than formerly married lone parents, but that this difference to a large extent reflects differences in the typical ages of these two groups (1998: 148).
Rowlingson and McKay also discovered that the rate of (re-)partnering varies according to housing situation (1998: 150). Other authors have also focused on lone parents and housing, sometimes because of the disadvantaged position of lone parents in the housing market (Kiernan et al. 1998), but sometimes with an emphasis on both constraints and choices (Crow and Hardey, in Hardey and Crow 1991). Clearly both the dynamics of transitions into and out of lone parenthood and also problematic housing are likely to generate actual or desired housing mobility among lone parents.
So far in this section an examination of the literature has highlighted the importance of the marital histories, work histories, and housing histories of lone parents. Among these various significant aspects of lone parents’ past, current and future lives it would be surprisingly easy to lose sight of perhaps the most significant type of history of them all, i.e. lone parents’ fertility histories. More specifically, while further child-bearing within lone parenthood is frequently stigmatised (Rowlingson and McKay 1998: 177), having further children may be an important feature of lone parents’ plans for the future (and may be linked to any plans that they have in relation to (re-)partnering.)
At this point it is perhaps worth noting that many of the recurrent themes in the literature about lone parents are of limited relevance to the question addressed by this project. The question highlights lone parents as social actors, planning and making decisions, whereas much of the literature addresses the structural situations of lone parents, e.g. as (marginalised) members of a society riddled with various family-related ideologies (McIntosh, in Silva 1996), as members of a possible ‘underclass’ (Roseneil and Mann, in Silva 1996), or as the focus of political or social policy agendas (Lewis 1997). Clearly, aspects of the structural situations of lone parents may constitute some of the factors which influence lone parents’ plans and expectations; however, while the effects of ideologies and social policies may be visible in some of the patterns revealed by the statistical analyses in this project, factors of this sort are not easily incorporated as variables within such analyses.
In conclusion, the literature examined within this section has revealed a number of significant dimensions of lone parents’ future lives with respect to which they might be expected to have plans and expectations, namely (paid) work, couple relationships, housing, and (further) children. In addition, gender, age and (past) marital status would appear to be potentially important sources of heterogeneity among lone parents.
A notable absentee from the various themes and characteristics identified above is social class. With the exception of references to the ‘underclass debate’, the literature examined seems to pay very little attention to class and class differences. This may reflect the fact that many lone mothers are not in paid work, and cannot therefore be allocated to social classes on the basis of their current occupations. However, it seems reasonable to hypothesise that social class (defined in some other fashion) might have an impact on lone parents’ plans and expectations, e.g. women in the social class containing professionals and managers might have less of an incentive to (re-)partner.
Data, methods and methodology
The source of data
This project is based on data from a survey commissioned and funded by the Department of Social Security (see Bradshaw and Millar 1991). The survey was designed to collect data from a representative sample of lone parents in the UK (excluding widows and widowers). While no comprehensive sampling frame exists for lone parents, it was felt that a reasonably comprehensive sampling frame could be assembled from records of families claiming one-parent benefit or income support. (Bradshaw and Millar do not, however, make it clear how many lone parents they think are outside this sampling frame). Lone parents were selected in April 1989 via cluster sampling based on DSS local offices. The survey’s response rate was 59%, largely reflecting migration and refusals (Bradshaw and Millar 1991: 3); comparisons of the sample with the sampling frame and with data from the General Household Survey highlighted some regional variations in response rates and a possible bias towards the inclusion of single lone mothers with younger children.