QUANTITATIVE DATA ANALYSIS AND THE RESEARCH PROCESS In Ulis section, (he way in which quantitative dam analysis fits into the research process - specifically the process of quantitative research - will be explored. . As we will see, the area covered by this book does not solely address the question of how to deal with quantitative data, since it is also concerned with other aspects of the research process that impinge on data analysis. Figure 1.1 provides an illustration of Hie chief steps in the process of quantitative research. Although Ihcre are grounds for doubling whelher research always conforms lo a neat linear sequence (Bryman 1988a, 1988b). [he components depicted in Figure I.I provide a useful model. The following stages are delineated by lite model. Theory The starting point for the process is a theoretical domain. Theories in (lie social sciences can vary between abstract general approaches (such as functionalism) and fairly low-level theories to explain specific phenomena (such as voting behaviour, delinquency, aggressiveness). By and large, the theories that are most likely to receive direel empirical attention are those which are at a fairly low level of generality. Merlon (1967) referred to these as theories of the middle range, lo denote theories that stood between general, abstract theories and empirical findings. Thus, Hirschi (1969), for example, formulated a 'control theory' of juvenile delinquency which proposes that delinquent acts arc more likely to occur when the child's bonds to society arc breached. This theory in large purl derived from other theories and also from research findings relating lo juvenile delinquency. Daca analysis and the research process 3 ---------'------------—*■ Theory l Hypolhesis Operationalizalion o( concepts Ý Selection of respondents or participants Survey/Correlational design Experimental design Conduct interviews i or administer questionnaires Create experimental iind conlrol groups Carry out observalions and/or administer tests or questionnaires Collect data Analyse data Findings ŕ Figure i.i The research process Hypothesis Once a theory has been formulated, it is likely thai researchers will want lo test it. Does the theory hold water when faced with empirical evidence? However, it is rarely possible to test a theory as such. Instead, we are more likely to find that a hypothesis, which relates lo a limited facet of the theory, will be deduced from the theory and submitted to a searching enquiry. For example, Hirschi, drawing upon his control theory, stipulates that children who are tied to conventional society (in the sense of adhering to conventional values and participating or aspiring to participate in conventional values) will be less likely to commit delinquent acts than those not so lied. Hypotheses very often take the form of relationships between two or more entities -- in this case commitment lo conventional society and juvenile delinquency. These 'entities' are usually referred lo as 'concepts'; that is. categories in which arc stored our ideas and observations about common elements in the world. The nature of concepts is discussed in greater detail in Chapter 4. Although hypotheses have the advantage that they force researchers to think systematically about what they want to study and to structure their research plans accordingly, they exhibit a potential 4 Data analym and the research process disadvantage in (hat (hey may divert a rcsearcher'.s attention 100 tar away from oilier interesting facts of tlie data l»c or she hau amassed. Operationalization of concepts In Order to USMH the validity nl a hypothesis ii is necessary to develop measures ol the constituent i nnccpts. This piocess is often referred to as Opgratiom lion, following expositions of the measurement pfOCOSS in physics (liridgman 1927). In effect, what is happening here is the translation of the concepts into v.iiiahlcs; dial is, attributes Ott which rcla.iul ub|ccls (individuals. Minis, nations, or whatever) differ. Ilin.chi operational i/ed the idea ol commitment to conventional society in a number of ways. One route was through a question on a questionnaire asking the children to whom it was to be administered whether they liked school. Delinquency was measured in one of two ways, of which one was to ask about t lie number of delinquent acts (o which children .»hunted (i.e. self-reported delinquent acts). In much experimental research in psychology, the measurement of concepts is achieved through (he observation Qf people, rather than through the admiuistiation of questionnaires. I U example, if ihr researcher IS interested in aggression, a laboratory siluation may be set up in which variations in ftggresstvc behaviour are observed. Another way in which t oneepttmay be npriaiionali/rd is thiough the analysis of existing statistics. ol which Din k heim \ (I '»52/1H9H) classic analysis of suicide rtltefl is an example A number of issues to do with the process of devising measures ol concepts and some of the properties thai measures should possess are discussed in Chapter 4. Selection of respondents or participants If a survey investigation is being undertaken, the researcher must fiud relevant people to whom the research instrument (hat has been devised (e.g. sc I f-administered questionnaire, interview schedule) should be administered. Hindu, for example, randomly selected over 5.500 schoohhddrcn from ,u\ area in California. The fact of random selection is important here because it reflects a commitment to the jiroduction .»I findings that can Ik- general i/cd beyond the confines of those who participate in a study. Il is rarely possible to contact all units in ,i population, so that a sample invariably lias 10 be selected. In order to be able to gcnciah/e to a widei population, a nprtstntativt sample, such u one that can he achieved through random sampling, will I* required. Moreover, many of tlw statistical techniques to he envcred in ibis book are iiif,>or4tory or in I held siluation. and observes »lie effects of thai manipulation on experimental participants There must also !»■ a 'control group' which ai i | | point o.......parison with ihe group of participants who icceive the experimental manipulation. With a survey/cornlatlonai design, die researcher does not manipulate any of the variables of interest and data relating to all vaiiables are collected simultaneously. The term correlation also refers to a technique for analyzing relationships between variables (see Chapter K), but is used in (he present context to denote a type of research design. The researcher does not always have a choice regarding which nf the two designs can be adopted. For example, llirschi could not make some children committed to school and olivets less committed and observe the effects on (heir propensity (o commit delinquent acts. Some variables, like most of those studied by sociologists, arc not ca(iable of manipulation. However, there arc areas of research in which top» I and hypotheses arc addressed wilh both types Of research design (e.g. the study of the effects of participation el work on job satisfaction and performance Me Bryman ľ>N6; Locke ami Schweiger l')70). It should be noted that m most cases, therefore, die nature o( die research design whethei experimental oi Mirvcy/coriel,ihuii.il is known .it the..... >i die sequeiu ■■ m ti ficd by Figure I.I. so tli.it research design characteristics permeate anil inform a numbei of stages of the research ptoccss. The nature of the research design has implications for (he kinds of statistical manipulation that can be performed on ihe resulting data, The differences between the two designs are given greater attention in (he next section. Collect data The researcher collects data at this stage, by interview, questionnaire, ohser-vation, or whatever. The technicalities of the issues pertinent to this .stage are not usually associated with a book such us this. Readers should consult a textbook concerned with social and psychological research methods it (hi untamiliai with the relevant issues, Analyze data This stage connects very directly with die material covered in this hook. At a minimum, the researcher is likely to want to describe his or her paiticip.uils in terms ol (hevariablcsdcrivingfrom the study. For example, Ihe rc.se.m bei might be interested in the proportion of children who claim to have committed no. 90 89 90 6 Daca analysis and tha research process just one, or two or mote delinquent acts. The various ways of analyzing and presenting the information relating to a single variable (sometimes called univariate analysis) are examined in Chapter 5. However, the analysis of a single variable is unlikely to suffice and the researcher will probably be interested in the connection between lhal variable and each of a number of olher variables. i.e. bivariate analysis. The examination of connections among variables can take either of two forms, A researcher who has conducted an experiment may be interested in ihc extent to which experimental and control groups differ in some respect. Tor example, the researcher might be interested in examining whether watching violent films increases aggressiveness. The experimental group (which watches the violent films) and the control group (which does not) can then be compared to sec how far they differ. The techniques for examining differences are explored in Chapter 7. The researcher may lie interested in relationships between variables - are two variables connected with each other so that they tend to vary together? For example, Hirschi (1969: 121) presents a tabic which shows how liking school and self-reported delinquent acts are interconnected. He found thai whereas only 9 per cent of children wlw say they like school have commuted two or more delinquent acts, 49 per .cut of those who say they dislike school have committed as many delinquent acts. The ways in which relationships among pnirs of variables can be elucidated can be found in Chapter 8. Very often (Ik researcher will be interested in exploring connections among more than two variables, i.e. multivariate analysis. Chapter 9 examines such analysis in the context of the exploration of differences, while Chapter 10 looks at the multivariate analysis of relationships among more than two variables. The distinction between studying differences and studying relationships is not always clear-cut. We might find that boys are mote likely than girls to commit delinquent acts. This finding could be taken to mean that boys and girls differ in terms of propensity to engage in delinquent ocis or lhat there is a relationship between gender and delinquency. Findings If the analysis of data suggests lhat a hypothesis is confirmed, this result can be ■led back into the theory that prompted it. Future researchers can then concern themselves either with seeking lo replicate the finding or with other ramifications of the theory. However, the refutation of a hypothesis can be just as important in that it may suggest that the theory is faulty, or al the very least in need of revision. Sometimes Ihc hypothesis may be confirmed in some respects only. For example, a multivariate analysis may suggest that a rclalionship between two variables pertains to only some members of a sample, hut not others (e.g. women but not men, or younger hul not older people). Such a lidding will require a reformulation of the theory. Not all findings will necessarily relate directly io a hypothesis. Willi a social survey, for example, the researcher may collect data on topics whose relevance only becomes evident at a later juncture. Data analysis and ih« ranirch process 7 As suggested above, lite sequence depicted in Figure 1 I <. onsiiiuics a model of the research process, which may not always be reproduced in reality. None the less, it does serve to pinpoint the importance to the process of quantitative research of developing measures of concepts and Ihc thorough analysis of subsequent data. One point thai was not mentioned in the discussion is the form lhat ilw hypotheses and findings lend to assume. One of ihc main aims of much quantitative research in ihc social sciences is the demonsiiation of causality -that one variable has an impact upon another. The terms independent variable am! dependent variable are often employed in this context. The former denotes a variable that has an impact upon the dependent variable. The latter, in other words, is deemed to be an effect of ihc independent variable. This causal imageiy is widespread in the social sciences and a major role of multivariate analysis is the elucidation of such causal relationships (Uryman 1988a). The ease with which a researcher can establish cause and effect relationships is strongly affected by the nature of the research design and il is (o this topic that we shall now turn.