Skip to main content Accessibility help
×
  1. Home
  2. Browse subjects
  3. Social Science Research Methods
  4. Qualitative research methods
  5. Content Listing

Refine search

Refine search

Access:
Content type:
Publication date:
Apply   From and To filters
Subject: Show more
Journals Show more
Publishers: Show more
Societies Show more
Series: Show more
Collections: Show more

Actions for selected content:

Select all | Deselect all
  • View selected items
  • Export citations
  • Download PDF (zip)
  • Save to Kindle
  • Save to Dropbox
  • Save to Google Drive

Save Search

You can save your searches here and later view and run them again in "My saved searches".

Please provide a title, maximum of 40 characters.
Cancel
×

1091 results in Qualitative research methods

Page 38 of 55

  • 9 - Looking for Differences Among Multiple Treatments

  • from Part III - Creating Experiments
  • Bernard C. Beins, Ithaca College, New York, Maureen A. McCarthy, Ball State University, Indiana
  • Book:
    Research Methods and Statistics
    Published online:
    06 August 2018
    Print publication:
    02 November 2017, pp 203-234

  • Summary

    CHAPTER PREVIEW

    In the previous chapters we introduced probability as the foundation for testing differences between two treatments. We then described how the independent t test allows us to determine if statistically significant differences are present between two groups receiving two different levels of an IV. Similarly, the repeated measures t test allows us to use a more sensitive analysis to examine differences in treatments using one group of subjects who experience both conditions. In this chapter we expand the probability model for testing statistical significance to include more than two levels of the IV. We explain how we can use an Analysis of Variance (ANOVA) to test for statistical significance when we have more than two groups or when participants experience more than two treatment conditions.

    A between groups ANOVA is an inferential test of probability that we can use with one independent variable having more than two levels. In addition to determining whether groups differ as a function of the level of the independent variable, we must conduct additional statistical tests to determine where, exactly, the differences exist. We conduct post hoc analyses to isolate specific differences when we have more than two groups. We also derive an overall measure of effect size as an estimate of the strength of the treatment effect.

    In most ways, a repeated measures ANOVA is much like the between groups ANOVA. The difference between these two inferential statistical tests is whether different groups receive the levels of the independent variable, or if one group of participants experiences all of the conditions, or levels, of the independent variable. The repeated measures ANOVA is more sensitive to detecting a statistically significant outcome because there is less error. So, when we find a statistically significant outcome, we also conduct post hoc analyses and calculate a measure of effect size.

    Statistical Testing for Multiple Treatments

    Thus far, our discussion of statistical analyses has been limited to comparing two levels of an independent variable. In the previous chapter, we used an example comparing two treatments for depression (i.e., treatment and control group), and we measured the effectiveness of the treatment using the Beck Depression Inventory (BDI-II). However, we frequently encounter research questions that include more than two levels of treatment.

  • 11 - Reading and Reviewing

  • from PART III - PROCESS AND PRESENTATION
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 169-181

  • Summary

    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?

  • 4 - Analyses

  • from PART I - BUILDING BLOCKS
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 47-62

  • Summary

    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.)

  • 18 - Univariate Statistics

  • from PART IV - STATISTICS
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 282-291

  • Summary

    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.

  • 23 - Causal Inference

  • from PART IV - STATISTICS
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 353-370

  • Summary

    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.

  • 22 - Regression

  • from PART IV - STATISTICS
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 331-352

  • Summary

    In this chapter we build on the concept of correlation with ordinary least squares (OLS) regression, even though the latter was invented first. If we think about our scatter plot, we could draw one line through the data that best fits all of our data points. Correlation is a statement about how close the points are to the line. The objective of regression is to determine the best fitting line for the data. Using regression, we can determine the average effect of our independent variable and make predictions about cases outside our sample. We will begin with the simplest regression model, bivariate, where the dependent variable is a function of a single independent variable, before expanding to consider multivariate models with multiple independent variables.

    Bivariate Regression

    As social scientists we primarily want to explain why variables of interest vary and vary together. Regression allows us the ability to measure the effect of one variable on another. It tells us the effect of an independent variable on a dependent variable. Furthermore, it provides us with the degree of the effect, thereby providing more explanatory leverage than in any other technique we have discussed thus far. Not unlike correlation we can find the strength and direction of association between two variables. Here, however, we can also get at the specific nature of the relationship; i.e., how much variance in the dependent variable is “explained” by the independent variable.

    In discussing correlation we implied without much specificity that one could draw a straight line that passed through the set of points in a manner that represented the overall pattern, positive or negative, steep or shallow. In addition to moving to thinking about causal relationships between independent and dependent variables (which was not required for correlation), with regression we also ask: Which linear relationship? In other words, of all the lines in Figure 22.1 that pass through the graph centroid – the point where X and Y intersect and marked in the figure by the intersection of the dotted lines – which fits the data the best?

    Before delving into the math, it is useful to graphically illustrate the characteristics of the best fitting regression line, as in Figure 22.2. For any line, we can measure the vertical distance between the line and each observation, which is called a residual.

  • 10 - Diverse Tools of Causal Inference

  • from PART II - CAUSALITY
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 156-166

  • Summary

    This section of the book has covered a lot of ground. We have presented a variety of causal frameworks (Chapter 5), defined the problem of causal inference (Chapter 6), and introduced an itinerary of research designs, which we classified as experimental (Chapter 7), large-N observational (Chapter 8), or case study (Chapter 9).

    This short chapter summarizes some of the themes discussed in this long, and complex, part of the book. We begin by reviewing the problem of confounding, and the various types of confounding that impede causal inference. Next, we review the panoply of research designs discussed in previous chapters, illustrating their interrelationship in a tree diagram, and discussing whether there is a “best” research design. We conclude by discussing the multi-method approach to research.

    Identifying and Avoiding Confounders

    Causal inference is possible only if confounders can be avoided or controlled in the analysis. Recall that we adopted a very broad definition of confounding, including any factor that produces a spurious (or biased) association between X and Y.

    This idea may be rendered in a more concrete fashion using the language of causal graphs. A causal model will generally render a valid estimate if the “frontdoor” path from X to Y is unblocked and any existing “backdoor” paths from X to Y are blocked. Simply stated, a causal model should allow no covariation between X and Y except that which is a product of X's causal impact on Y. This is another way of describing causal comparability, i.e., the expected value of Y for a given value of X is the same for all units throughout the period of analysis. (You will see that there are many languages by which to express the same general idea.)

    Five types of confounders, introduced in previous chapters, are recapitulated in Figure 10.1. In each case, Z stands for the confounder, generating a backdoor path from X to Y. Recall that in interpreting these graphs one must distinguish settings in which Z is conditioned from settings in which it is not (indicated by square brackets). In some settings, a confounder is created because a factor is not conditioned, in other settings because it is conditioned.

  • 12 - Brainstorming

  • from PART III - PROCESS AND PRESENTATION
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 182-192

  • Summary

    How does one go about identifying a fruitful research question and, ultimately, a specific research hypothesis? This is the very early exploratory phase, when one quite literally does not know what one is looking for, or at. Arguably, it is the most crucial stage of all. Nothing of interest is likely to emanate from research on topics that are trivial, redundant, or theoretically bland – no matter how strong its research design components are.

    Methodologists generally leave this task to the realm of metaphor – bells, brainstorms, dreams, flashes, impregnations, light bulbs, showers, sparks, and whatnot. The reason for this lack of attention is perhaps because beginnings are inherently unformulaic. There are few rules or criteria for uncovering new questions or new hypotheses. Methodologists may feel that there is nothing – nothing scientific, at any rate – that they can say about this process. Karl Popper states the matter forthrightly. “There is no such thing as a logical method of having new ideas,” he writes. “Discovery contains ‘an irrational element,’ or a ‘creative intuition.’”

    Accordingly, what we have to offer in this chapter is more in the character of a homily than a framework. It reads like an advice column. We urge you to Study the Tradition, Begin Where You Are, Get Off Your Home Turf, Play With Ideas, Practice Disbelief, Observe Empathically, Theorize Wildly, and Think Ahead.

    This advice is largely commonsensical and by no means comprehensive. It cannot help but reflect our own views and experiences, though we have drawn extensively on the writings of other scholars. Nonetheless, it may help to orient those who are setting out on their first journey, or who wish to begin again.

    Study the Tradition

    Consider the state of the field on a topic. What are the frontiers of knowledge? What do we – collectively, as a discipline – know, and what don't we know? We doubt if anyone has happened upon a really interesting research topic simply by reading a review of the extant literature. However, this is an efficient method of determining where the state of a field lies and where it might be headed.

  • 1 - A Unified Framework

  • from PART I - BUILDING BLOCKS
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 3-13

  • Summary

    The purpose of social science is to make a difference in the world by applying reason and evidence to problems of general concern. Every question of social science relates (or ought to relate) to normative concepts such as justice, democracy, human rights, prosperity, happiness, or equality.

    What distinguishes social science from casual conversation, journalism, or political rhetoric may be summarized as follows. First, social science involves the systematic application of reason and evidence to problems with explicit attention to method and to possible sources of error. Second, social science is accompanied by realistic estimates of uncertainty with respect to whatever conclusions are drawn from the evidence. Third, social science attempts to provide a comprehensive treatment of a subject within whatever scope-conditions are defined by the study. All relevant information should be included; none should be arbitrarily excluded. Finally, social science adopts a disinterested posture with respect to all goals except the truth. Its purpose is to get as close to the truth as possible, in all its complexity, rather than to provoke, entertain, elucidate moral truths, or advance polemical claims.

    These features render social science less stimulating than other media, where there is generally a premium on brevity, accessibility, provocation, righteousness, or humor. Social science is a sober profession. However, for those excited by the prospect of getting it right, and willing to expend some energy to get there, the practice of social science may be highly rewarding.

    Consider the problem of crime, a topic that often evokes hot rhetoric and strong opinions. Most media reports and political speeches offer little useful information about the prevalence of crime, its sources, and its potential solutions. Instead, they exploit the public's fascination with gruesome events and, in the process, provoke fear. From this perspective, the cold gaze of social science offers some relief.

    Researchers have spent a good deal of time studying the rise and fall of violent crime in the United States and elsewhere. In the early 1960s, the United States enjoyed a low homicide rate of 5 murders per 100,000 inhabitants. Over the next two decades this rate doubled – to 10 per 100,000 inhabitants – peaking in the late 1970s or early 1980s, at which point the United States could claim the highest rate of violent crime of any advanced industrial country.

  • 9 - Case Study Designs

  • from PART II - CAUSALITY
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 139-155

  • Summary

    There are two ways to learn about a subject. One might study many examples at once, focusing on a few selected dimensions of the phenomena. We shall refer to this as an extensive approach, as laid out in Chapters 7–8.

    Alternatively, one might study a particular example, or several examples, in greater depth. We shall refer to this as an intensive, or case study, approach – the topic of this chapter.

    A case connotes a spatially delimited phenomenon (a unit) observed at a single point in time or over some period of time. It may be a political unit with a defined area of semi-sovereignty (e.g., empire, nation-state, region, municipality), organization (e.g., firm, non-governmental organization, political party, school), social group (as defined, e.g., by ethnicity, race, age, class, gender, or sexuality), event (e.g., foreign policy crisis, revolution, democratic transition, decision-point), or individual (e.g., a biography, case history).

    However defined, a case must comprise the type of phenomenon that an argument attempts to describe or explain. In a study about nation-states cases are comprised of nation-states (observed over time). In a study that attempts to explain the behavior of individuals, cases are comprised of individuals. And so forth.

    A case study research design is an intensive study of a single case or a small number of cases that promises to shed light on a larger population of cases. The individual case(s) is viewed as a case of something broader, just as large-sample analysis is also generally viewed as exemplary of a broader phenomenon. Thus, both intensive and extensive analyses generally make inferences from a sample to a population, even though the sample sizes are very different.

    Case study research may incorporate one or several cases. The latter is a defining characteristic of comparative historical analysis, associated with the work of Barrington Moore, Theda Skocpol, David Collier, and James Mahoney. However, as the sample of cases expands it becomes less and less feasible to investigate each case intensively. The case study format is thus implicitly a small-sample format.

  • 13 - Data Gathering

  • from PART III - PROCESS AND PRESENTATION
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 193-229

  • Summary

    Insofar as social science research rests on an empirical base it involves an encounter with data. Data may be qualitative or quantitative. It may be numerical, textual, visual, olfactory, auditory, or artifactual. It may be drawn from primary or secondary sources. It may be collected by the researcher or by someone else. “Data” is understood here in the broadest sense, including any sort of evidence that serves as the empirical basis for understanding and explanation.

    In this chapter we lay out various methods of gathering data, a topic that is important for both producers and consumers of social science. Note that in order to judge the quality of a study you need to be familiar with the data-collection method that underlies the analysis.

    Data-gathering methods may be classified broadly as obtrusive or unobtrusive. Obtrusive methods comprise standardized surveys as well as less structured interpersonal settings such as interviews, focus groups, and ethnography. Unobtrusive measures include those that are surreptitious and ex post. After outlining these techniques of data collection we discuss the problem of data assessment: how do we know that our data (or someone else's) is true?

    Before launching into the chapter, a few caveats and clarifications are in order.

    First, each section of the chapter is cumulative. For example, many of the ingredients of a good survey are also ingredients of a good interview, focus group, or ethnography. This explains why the section on surveys is somewhat longer than the later sections of the chapter. It also means that this chapter should probably be read from front to back rather than browsed in a selective fashion.

    Second, each method will be treated as a stand-alone technique of data collection. However, readers should be aware that in the course of conducting a study it is common to combine several data-collection strategies. Where a multi-method approach is adopted, the strengths and weaknesses of each strategy should be understood within that context.

    Third, experiments and case studies are regarded in this book as research designs rather than data-gathering techniques and are therefore discussed in Part II rather than in this chapter.

  • 21 - Bivariate Statistics

  • from PART IV - STATISTICS
  • John Gerring, University of Texas, Austin, Dino Christenson, Boston University
  • Book:
    Applied Social Science Methodology
    Published online:
    28 May 2018
    Print publication:
    24 April 2017, pp 313-330

  • Summary

    Bivariate statistics allow us to test the relationship between two variables. While simple, they provide great empirical leverage for hypotheses of association and, with the appropriate research design, causality. We next hone our ability to make controlled comparisons and introduce inference making about sample means with the difference of means test. We then proceed to correlation, which moves us beyond making a simple claim of a relationship or no relationship between two variables to a measure of both the strength and direction of the relationship.

    Revisiting Levels of Measurement

    In the previous chapter, we found out some important substantive information about our population means. Indeed we learned how to estimate a population mean from the information that we gain in a single sample. However, we are still not in a position to assess a hypothesis. We do not generally have expectations as simple as “the mean number of parties in a democracy is x.” Our hypotheses typically suggest a relationship between an independent and dependent variable, and we have not made reference to bivariate analyses yet. We will explore a number of statistical tests for assessing hypotheses. We can divide hypothesis tests into three types based on the information they provide about the relationship between the independent and dependent variable:

  • 1. Those that simply analyze whether or not there is a relationship between variables (e.g., Difference of Means).

  • 2. Measures of association, which tell us the direction and strength of the relationship between two variables. (e.g., Correlation).

  • 3. Measures of average effect, which tell us the amount of change in the dependent variable given a unit change in the independent variable (e.g., Regression).

    • It is important to note that knowing which test to use requires us to think not just about the question we would like to answer but also our variables’ levels of measurement, their population distributions, and the sample size. In addition, all the tests discussed here demand that the data come from a random sample. In Chapter 3 we learned about levels of measurement. In quantitative analyses understanding the level of measurement is essential because it helps us decide which hypothesis test to use.

Page 38 of 55