This chapter is devoted to reading and reviewing the social science literature on a topic. Here, we discuss how to distinguish social science sources from other sorts of work, how to locate and survey the literature on a chosen topic, how to read strategically, how to read critically, how to figure out complex arguments, how to construct a systematic review of the literature on a topic, and how to take notes as you go along. These are closely linked topics so you will find a good deal of overlap across these sections.

As you read social science you may be struck by the dry tone of the literature, especially in journal articles. (Books usually attempt to include some divertissement to relieve the tedium of pages and pages of prose.) Remind yourself that what is exciting about science – any science – is getting closer to the truth. The mode of exposition is secondary, and generally remains in the background. Indeed, it is important to adopt a dispassionate tone to discourage other factors from interfering with the logic of the argument. This is why authors generally avoid personal anecdotes, jokes, heavily symbolic or allegorical language, and other narrative devices.

Of course, there is a person behind the prose, and perhaps it would aid the cause of science if some personal elements – such as his/her motivations for studying a subject – were made explicit. But, for better or for worse, this is not the accepted scientific mode of communication in most social science fields. Bear with it.

In any case, new media are opening up new modes of communication, many of which are more personal in nature and less tightly structured. On virtually any given topic you can now find blogs (discussed below) or lectures and debates preserved on YouTube, as well as other multi-media presentations. The medium is changing, though the austere format of journal articles is likely to remain the workhorse of social science for the foreseeable future.

Identifying Social Science Sources

Reading social science presumes that one can identify works of social science from the mass of other sources out there in print and on the World Wide Web. This brings us back to an earlier discussion. What is it that distinguishes social science from other genres such as casual conversation, journalism, or partisan rhetoric?

Introduction

The classroom is the core unit of our educational system, and it illustrates many common social-psychological concepts and phenomena. Whenever people come together for the purpose of learning, social-psychological principles may be applied to further educational goals. Applying these principles to the classroom may help increase students’ commitment to learning, make their attitudes towards school more positive, reduce their feeling of failure and related negative effects or emotions, and improve their level of aspiration as well as their grades. Clearly, education is – at least in part – applied social psychology. Indeed, in the past three decades, a new scientific area has emerged, the so-called ‘social psychology of education’, ‘educational social psychology’, or sometimes even ‘social educational psychology’. According to Feldman (1986), this area represents an amalgamation of social psychology and education, and has produced a broad range of theories, research, and data that speak to the interests of both educators and psychologists.

Our purpose here is not to describe all the current themes of interest that characterize this interface, but rather to focus on a limited set of social-psychological concepts and phenomena in relation to major processes in the classroom. First, we focus on the role of comparisons that students are constantly making with each other. Today, there is ample evidence that social comparisons contribute to students’ academic achievements, and we examine here the meaning and consequences of such comparisons in the classroom. Second, we focus on students’ ‘theories’ of intelligence. Over time, students come to believe that their performance in a wide range of tasks is or is not narrowly constrained by innate attributes, and that they can or cannot change their level of intelligence. We describe how such beliefs may affect students’ academic achievement. Third, we focus on students’ academic self-concept. Over time, students develop positive or negative views about their abilities in such-and-such an academic domain (e.g., math), which become an integral part of their ‘academic self-concept’. Does this self-concept also make a difference, and how? Fourth, we focus on social stereotypes. Students belong to social groups and may sometimes suffer from the negative stereotypes associated with these groups. Are these negative stereotypes a real problem for students, and why? Finally, we focus on the ‘classroom climate’.

We began (in Chapter 2) with arguments and proceeded (in Chapter 3) to conceptualization and measurement. In this chapter, we turn to the problem of how to analyze an argument empirically. This may be referred to variously as appraisal, assessment, corroboration, demonstration, empirics, evaluation, methods, proof, or testing. Pursued in a self-conscious fashion, this stage of research involves a research design, i.e., an explicit method of selecting and analyzing data.

We begin by introducing a set of key terms that are necessary to understand the construction of a research design. We proceed to a discussion of the general issues that all analyses encounter. This includes precision and validity, internal and external validity, sample representativeness, sample size, probability and non-probability sampling, and missing-ness. The terms and topics introduced in this chapter will enter the narrative in later chapters repeatedly. This chapter therefore plays a foundational role in the book.

Definitions

A standard empirical analysis involves a number of components, which must be clarified before we continue. Much of this vocabulary is borrowed from survey research. Nonetheless, the concepts are helpful in all styles of research, whether quantitative or qualitative, and are illustrated in Figure 4.1.

The most basic unit in any analysis is an observation. Observations are the pieces of evidence deemed relevant for demonstrating an argument. In a standard matrix (rectangular) dataset, an observation is usually represented as a row. Each row in Figure 4.1 represents a single observation.

Each observation should record values for all relevant variables. In causal analysis, this includes X (the causal factor of theoretical interest) and Y (the outcome of interest), along with any other variables deemed essential for the analysis. In a rectangular dataset, variables are usually represented with vertical lines. There are three variables in Figure 4.1: X, Z, and Y.

An observation is drawn from a unit or case – bounded entities such as individuals, organizations, communities, or nation-states, which may be observed spatially and/or temporally (through time). The terms unit and case are more or less equivalent. (While a unit is bounded spatially, a case may also have implicit or explicit temporal boundaries.)

The purpose of this chapter is to show how we can parsimoniously describe data so that we may share our understandings of it with others. Even before calculating statistics we can begin to explore the nature of variables simply by presenting them in a particular manner. A host of both tabular and graphical presentations are available to describe data. We will touch on a few of the most helpful visualizations and then turn to univariate statistics. Univariate statistics restricts the analysis to a single variable. It typically concerns the variable's level of measurement, range of values, and the average or central tendency.

Frequency Distributions, Bar Graphs, and Histograms

We begin with the question of how to explain a lot of data with fewer numbers and fewer words. That is, ask yourself how you might quickly explain to somebody else who does not have a deep knowledge of the data the basic features of it. The answer is found in tables, graphics, and statistics. In each of these broad categories are a host of techniques for reducing the data down to a more manageable collection or visualization.

Think about how a tabular summary of data might more succinctly describe data than looking over the full dataset. Frequency distributions present the number of occurrences of an event of interest in a table. The data are grouped into mutually exclusive categories and summarized to show the number of occurrences in each category. The format of the table includes a title, clearly describing the concept, as well as columns referring to the values, categories, and frequency of categories. Typically, the first column lists the numeric values of a variable and the second the categories to which the values refer. The next column lists the number of units in each category, or the frequency, and the total number of observations in the margins.

Frequency distribution tables often include additional information about the distribution, including the cumulative frequency and relative frequency of the categories. The cumulative frequency shows the sum of the frequencies for the previous categories. Thus it is important in frequency distributions to arrange the categories either from top down or bottom up when they are ordered. The relative frequency is the fraction or proportion of times an outcome in a category occurs out of the total number of possible outcomes.

Everyone has opinions about mass education. Not only did we all go to school – so we know what it's like from the inside – but also we hear about educational issues and concerns in the news every day. These might include questions such as, should we go back to basics when teaching literacy? Should we reduce funding to state schools that do poorly on standardised testing? Should we teach more about the environment? Our opinions on these specific matters, assuming we have any, tend to be shaped by our more general beliefs about education – schools aren't educating our children properly; our education system seems pretty fair; the modern curriculum is full of trendy rubbish. The problem here is that many of these things we think we know about our system of mass education don't stand up to any kind of close scrutiny; they are often myths, and this book will address the most important of these myths.

This demythologising approach is unusual, but by no means unheard of, within social analysis. One of the greatest of all writers in the field, Peter Berger, suggests in his seminal text Invitation to Sociology (1963, 51) that ‘debunking’ is one of the most crucial elements of the discipline, and that there is ‘a logical imperative to unmask the pretentions and the propaganda’ by which people cloak their actions with each other. To put it another way, he proposes that our social world is full of stories we tell ourselves, and it is a sociologist's job to see which ones hold (at least some) water, and which ones don't.

However, not only is debunking/demythologising one of the central responsibilities of the social analyst, it is also an excellent pedagogic approach to the discipline. LeMoyne and Davis (2011) contend that the most effective way of coming to grips with complex social problems is to begin by examining important taken-for-granted cultural beliefs, and from there these beliefs can be systematically questioned – by illustrating the ways in which those beliefs evolved, by unpacking their theoretical domain assumptions, by providing concrete evidence that speaks to these beliefs, and by providing a vocabulary of other ways of seeing the same issues.

Much of social science research is concerned with causal relationships. In this chapter we explore the general framework for making causal statements using different methodological approaches. With experiments as our point of reference, we revisit regression in the context of a causal treatment variable, and then introduce a technique to evaluate the causal effect of a treatment with observational data by matching treated and control units. Before doing so we lay out the specific assumptions necessary for causality and the different motivating factors for each causal model.

Assumptions and Assignment

Why causal inference? Our statistical objective thus far has been to infer associations among variables. From these associations we can estimate probabilities of events with the statistical methods introduced above, provided that the external conditions remain the same. This allows us to answer important questions, like: What is the mean number of parties in democracies? Are turnout and geographic regions related? Are greater issue dimensions associated with greater numbers of parties? On the contrary, we need causal inference when we would like to infer probabilities under different conditions. That is, when we would like to know what would happen if something else happened. When we expect probabilities to change in response to external factors we must rely on causal analysis. Thus, causal questions ask somewhat different questions: What are the effects of worker-training programs? Does smoking cause cancer? Does viewing a campaign advertisement change vote preferences? In each case we are asking a question that posits different probabilities under different conditions: attending versus not attending a worker-training program; smoking versus not smoking; viewing versus not viewing a campaign ad.

Throughout the statistics section of this book we have identified hypothesis tests that are based on covariance between suspected cause and effect. However, the tests themselves are only covariational; that is, they are not explicitly causal. As we discussed in Chapters 21 and 22, in order to make claims of causality we need more than the evidence of covariation between cause and effect that we can attain from these statistical methods. Minimally, we also need the cause to precede the effect and to be able to eliminate plausible alternative causes.

Open any book on sociology, whether it is about education or not, and you can almost guarantee that there will be chapters on social class, gender and race. There are also likely to be other common themes – the family, ageing, deviance, and so on – but none of them is regarded as quite as foundational as the three familiar pillars of social structure. That is, not everyone has a family, or is old, or deviant, but everyone is deemed to have a social class, a gender and a race/ethnicity, and, more often than not, where you stand in relation to these categories correlates fairly directly with access to power, resources… and education. This part of the book will examine these three social axes and draw some conclusions about how they structure our society, and how they affect the probability of success within our schools.

Though these conclusions may still be interesting and informative, sociological analysis has moved on somewhat from the belief that there are three straightforward categories that shape the structure of our society, categories that have an objective reality that we can come to know – i.e. the singular ‘truth’ of social class, gender and race/ethnicity. While this modernist understanding has had some interesting things to tell us, it has largely been superseded by more postmodern approaches to the same issues. These approaches attempt to avoid the sweeping generalisations, and the either/or logic, of previous work in the area. As such – while staying on the same terrain as social class, gender and race/ethnicity – this section will also examine how these three solidly nineteenth and twentieth-century concepts can be best understood in the more nuanced sociological environment of the twenty-first century.

Chapter 1 – Social Class – examines the relationship between the first of the three pillars and our education system. It begins by questioning the common assertion that the United Kingdom is no longer a particularly unequal society; it also questions the belief that how well we do in school is determined by our intellect, and not our social backgrounds; finally, it challenges the assumption that social class is determined solely by how rich you are.

Introduction

According to some definitions, consumer behaviour consists of “the psychological and social processes people undergo in the acquisition, use, and disposal of products” (Bagozzi, Gürhan-Zanli, and Priester, 2002, p. 1). This seems a plausible definition at first. But it is deficient, in so far as it is difficult to find instances of behaviour that do not involve some goods. Sleeping comes to mind, but unless one sleeps naked and in the forest, it involves a bed and pyjamas, both probably branded goods. And even if we were to write this chapter with pens rather than computers, writing would still involve use of products: pens do not grow on trees, nor does paper. We therefore narrow the focus of this chapter on what is arguably the most characteristic aspect of consumer behaviour, namely the process of acquiring goods. Getting insight into the process of acquiring goods is not only interesting for the manufacturer trying to sell a new type of phone, but also for the organization trying to fight consumer overspending, or increase the relative sales of pro-environmental products (see Chapter 9), or healthy food (see Chapter 10). So what determines the likelihood that we acquire a specific good? To answer this question we first discuss the potential functions of goods for consumers. In the second part we look at two different types of buying behaviour: deliberate and impulse buying. We discuss which factors influence deliberate purchases and when impulse buying is more likely. In the third and final part we reflect on the impact of advertising on acquiring goods, as we will discuss the different routes through which advertising can influence consumers.

The function of goods

Consumers buy goods as a means to achieve some goal. Goals are desired outcomes that are also perceived as attainable. In a narrow utilitarian sense, consumers buy goods or services simply for the function they are supposed to serve (e.g., wine is for drinking and clothes are for wearing). However, consumers buy most goods not just for this utilitarian function, but also to communicate something about themselves to others and themselves; they serve an identity function (Govers and Schoormans, 2005; McCracken, 1986).

Introduction

Currently, 1.2 billion people live in extreme poverty worldwide. They suffer severely from hunger and lack of material possessions, and live on less than US$1.25 per day (World Bank, 2013). Large sums of money are spent on development aid projects to improve the living conditions of people in low income countries. In 2013, governments provided the highest amount ever to development aid, a total of US$134.8bn (OECD, 2014). This aid aims to support the economic, environmental, social, and political development in low income countries. For example, cooking stoves are distributed to decrease health problems, laptops are given to children to improve educational outcomes, or microcredits are provided so that people can develop their own income-generating activity. These examples all focus on improving people's living conditions in the long run.

To date, it is heavily debated whether investments in development aid can indeed improve the life of people in low income countries. One position, for example, argues that initial large investments are crucial to reduce poverty because they can help these countries to overcome their ‘poor’ position (i.e. they are hot, malaria infested, infertile, and landlocked; Sachs, 2005). Others (Moyo, 2005) argue that development aid worsens the situation because the way aid is distributed in these countries corrupts and undermines local institutions and creates powerful aid agencies.

Despite this discussion on the effects of investments in development aid in general, it is important to gain insights into how specific development aid interventions can change people's behaviour and achieve its goals. How can we study the effectiveness of such interventions? Which factors can increase its effectiveness? Does development aid have important (un)wanted side-effects? Social psychological theories and methodologies can play an important role in answering these questions. Social psychologists can help to gain a better understanding of when and how development aid projects are successful or unsuccessful in reaching their goals, by taking into account social psychological factors (see Chapter 1) and cultural differences. The first part of this chapter explains the importance of clearly formulating the primary goals and expected impacts of a specific development aid project and how the effectiveness of projects can be evaluated.

Introduction

Economic theory of utility maximization

The British philosopher Jeremy Bentham postulated in his book The Principles of Morals and Legislation (1789) that the pursuit of pleasure and the avoidance of pain explain all human behaviour. Later this principle was called ‘hedonic utility.’ Utility was then considered to be a cardinal measure of pleasure. Cardinal means that utility should be measured in absolute numbers. Disutility referred to negative emotions that people avoid such as pain. People select the maximum utility, the net balance of the pursuit of pleasure and the avoidance of pain. People act rationally if they maximize their utility. The notions of maximizing utility, and thus acting rationally, were extremely relevant for nineteenth century economists. Economists (Jevons, 1871/1970) saw the principle of hedonic calculus, the computation of utility as the net balance of pleasure and pain, as a universal law of human behaviour from which economic relations and market exchange behaviours could be derived.

Economists did not get the support of psychologists of that time for their principle of hedonic calculus. William James (1890) criticized this approach, arguing that the determinants of behaviour cannot be reduced to a simple hedonic calculus. This instigated the idea of freeing economic theory from psychological influences, either by accepting hedonic calculus as a simple starting point for economic theorizing, or by excluding psychology completely by measuring utility from expenditure or preference. It also implied that economists gradually shifted from measuring cardinal (metric) to ordinal (non-metric) utility. Ordinal utility is easier to measure, because it is based on ranking options. If a person prefers object A over object B, object A has a higher utility than object B, not quantifying how much higher. Cardinal utility is based on rating options on a utility scale. A person gives a utility score for each option. Economics started to build a theory of economic behaviour in terms of ordinal utility, free of any explicit psychological basis.

The road taken in economics to exclude psychological assumptions is comparable to the behaviouristic approach in psychology. In the behaviouristic approach, unobservable mental constructs are avoided as an explanation of behaviour. Only behavioural observations and causations are allowed in the explanation of behaviour.

Introduction

In the Netherlands, with a population of roughly 16 million people, around 1.2 million offences are registered by the police every year. Many of those crimes are committed by young people. In 2009, over 25 per cent of all suspects arrested were aged between 16 and 22. This is over 2.5 times as many as would be expected if age and (suspicion of a) crime were unrelated (Eggen and Kessels, 2010). Unsurprisingly, adolescent delinquency is high on the agenda of criminologists and policy makers alike.

Figure 13.1 depicts the age distribution of crime in 2009 based on Dutch police data. The graph shows a steep rise from age 12 onwards to a peak during the late teenage years when offending is most common. After that, the number of offenders gradually declines during the adult years to almost zero at old age. In criminology, this skewed bell-shaped association between age and crime is referred to as the age-crime curve, which has been replicated across many countries and historical periods. Explanations offered for the age-crime curve have differed radically, as have the remedies to combat delinquency and crime emanating from them.

Against the background of this age-crime debate, this chapter focuses on one of the currently most influential explanations of the association between age and crime: the Dual Taxonomy (Moffitt, 1993). This theoretical framework proposes a complementary set of developmental theories explaining patterns of both Life course persistent and Adolescence-limited offending which it is suggested underlie the age-crime curve. The chapter describes the effects of social cognition and social influence on criminal behaviour that figure prominently in this taxonomy, the way the core premises of the taxonomy have been tested, and prevention and intervention efforts that can be derived from this taxonomy.

Explaining the age-crime curve

Criminologists define the criminal career as the longitudinal sequence of crimes committed by an individual offender (Blumstein et al., 1986). Several criminal career dimensions can be distinguished, including (age of) onset or first-time participation in offending, the frequency of offending, and the duration of the offending period, defined as the interval between the first and last offence.

This chapter address one of the most important areas of philosophy – ethics – and uses it to examine aspects of the role of the law within education. Of all the areas of philosophy, more has probably been written about ethics, and over a longer period, than any other. In addition, all cultures are structured around a fundamental ethical system: the law. However, irrespective of their importance, both subjects are currently notable for their lowly status within the teacher education curriculum.

Research designs in applied social psychology

In 1999, Eric Harris and Dylan Klebold launched an assault on Columbine High School in Littleton, Colorado, murdering thirteen people and wounding twenty-three before killing themselves. Similar school shootings have occurred in other communities, such as Paducah, Kentucky, Jonesboro, Arkansas, Blacksburg, Virginia, and more recently at Sandy Hook elementary school in Connecticut. Although the exact causes for these attacks may never be known, one characteristic these events share is that all the shooters were young adults who habitually played violent video games. In light of this fact, many American politicians, including Senator Joe Lieberman and Vice President Joe Biden, responded to the recent Sandy Hook shooting by calling for a reassessment of the video game industry.

Since their introduction in the 1980s, video games have become increasingly popular. At the time, American children played video games an average of 4 hours per week (Harris and Williams, 1985). Now they play an average of 13 hours a week, with young boys averaging 16–18 hours (Martin and Oppenheim, 2007). Unfortunately, as video games increased in popularity, they also increased in violence. It is estimated that over 90 per cent of video games rated by the industry as ‘appropriate for everyone’ (E10+) contains violent content (Gentile, 2008). Despite the prevalence of violent video game content, little is known about the consequences of such exposure. A recent online poll suggests that only 13 per cent of people think video games play a role in school shootings (Goad, 2013), but just because people believe video games are harmless does not necessarily mean they are. Luckily, researcher Craig Anderson has devoted his career to this important social issue by designing a programme of research that investigates the relation between violent video game use and aggression. In our opinion, this line of research represents an ideal for the applied social scientist. This is because Anderson and his colleagues created a research programme that uses multiple studies with multiple designs in order to investigate a socially relevant topic. Our chapter will use this work as a springboard for discussing research design in applied social psychology.