Cross Sectional Research

6 Key excerpts on "Cross Sectional Research"

• 

  eBook - ePub

  Introducing Social Research Methods

  Essentials for Getting the Edge

  • Janet M. Ruane(Author)
  • 2015(Publication Date)
  • Wiley-Blackwell

    (Publisher)

  Consequently, cross-sectional research is quite common in social research. Obtaining information from a cross-section of a population at a single point in time is a reasonable strategy for pursuing many descriptive and exploratory research projects (see Chapter 2). Where do Americans currently stand on immigration reform? How do today’s youth feel about marriage? What are Baby Boomers’ thoughts on retirement? How do the people of the world regard Edward Snowden? Since the answers to these questions are about “now”, they all can be addressed via a cross-sectional design employing a survey: Indeed, until 2010, the cross-sectional design was the heart of the General Social Survey (GSS) conducted by the National Opinion Research Center (NORC) at the University of Chicago – a major research effort that documents a large array of Americans’ attitudes and behaviors. Through 2010, the GSS interviewed a cross-section of Americans every year or two in order to learn their views on a variety of topics. While it takes the GSS approximately two months to complete a surveys, it was still considered a single point in time study design (i.e. each respondent is only contacted once). This slice into time and across the population provided an extremely valuable and timely look at what Americans are currently thinking and doing. (In 2008, the GSS started transitioning to a panel design which was fully implemented in 2010 – see an explanation of this in section “Fixed-Sample Panel Designs”.) The United States is not alone in relying on massive cross-sectional studies to generate the most up-to-date “picture” of society. Australia, Britain, Canada, Germany, and many more nations around the world all have a vested interest in conducting some form of a general survey of its people. 1 Cross-sectional design is also useful for pursuing correlational analysis – that is, answering questions about associations between variables – a mainstay of much sociological research

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Research Design in Social Research

  • David de Vaus(Author)
  • 2001(Publication Date)
  • SAGE Publications Ltd

    (Publisher)

  11 Issues in Cross-Sectional Design

  Cross-sectional designs are probably the most widely used designs in social research. One reason for this popularity is that they enable the researcher to obtain results relatively quickly. Since data are collected at one point of time there is no need to wait for various follow-up stages or interventions before analysing the data. It is also true that, other things being equal (e.g. sample size, population, sample type), cross-sectional designs are more cost effective than comparable experimental and longitudinal designs. This is because cross-sectional designs do not entail the cost of repeated data collections, tracking respondents or of experimental interventions.

  Cross-sectional designs can be ideal for descriptive analysis. If we simply want to describe the characteristics of a population, their attitudes, their voting intention or their buying patterns then the crosssectional survey is a most satisfactory way of obtaining this descriptive information. But cross-sectional designs are not restricted to descriptive analysis. As will be argued below, proper analysis that uses statistical controls enables cross-sectional data to provide valuable information about causal processes and for testing causal models.

  There are, however, a number of methodological and practical issues of which we need to be aware when using cross-sectional designs. An awareness of these issues should help minimize the shortcomings of this design.

  Methodological Issues

  Internal Validity

  We encounter problems with internal validity when the logic and structure of the design does not enable us to choose unambiguously one explanation of our results over another explanation. Campbell and Stanley (1963) have identified a number of factors that threaten internal validity and these have been discussed in detail in Chapters 5 and 8

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Key Concepts in Health Psychology

  • Ian Albery, Marcus Munafo(Authors)
  • 2008(Publication Date)
  • SAGE Publications Ltd

    (Publisher)

  One of the most common and well-known study designs is the cross-sectional study design. In a cross-sectional study participants are observed only once, offering a ‘snapshot’ of the characteristics of interest at that particular moment. In this study design, a population, or a sample thereof, is studied in a single instance, either by examining a single well-defined group of individuals, or by comparing cases (namely, those with an identifiable clinical condition) versus controls (namely, healthy individuals).

  ORIGINS

  Cross-sectional study designs are widely employed in epidemiological research, where characteristics of individuals (e.g. those presenting with a particular disease) are related to potential risk factors, or characteristics of cases (e.g. those with a particular disease) compared to the characteristics of controls (usually healthy individuals drawn from the general population). An example might be that cholesterol levels are higher among patients with coronary heart disease compared to healthy controls. Cross-sectional studies tend to be faster and cheaper to conduct than longitudinal studies, for the obvious reason that participants need only be studied at a single time point, rather than repeatedly over several time points.

  CURRENT USAGE

  It may be tempting to interpret the results of a cross-sectional study as if they were drawn from a longitudinal study. For example, if we look at cholesterol levels and the proportion of dietary calories derived from saturated fat in a cross-sectional study of healthy individuals, we might expect to find a very strong correlation. In this case, one might be tempted to conclude that high levels of saturated fat in one’s diet cause increased cholesterol levels. This would be an example of treating cross-sectional data as longitudinal data, because one cannot make inferences about causation from cross-sectional data, however tempting it might be to do so (and even if one turned out to be correct!). A trivial example illustrates this: height and weight are very highly correlated, but it doesn’t make sense to suggest that height causes

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Introductory Statistics for Health and Nursing Using SPSS

  • Louise Marston(Author)
  • 2009(Publication Date)
  • SAGE Publications Ltd

    (Publisher)

  Chapter 7 .

  In cross sectional studies data can be collected about the past, but no data were collected prior to the encounter with the researcher. If questions that rely on memory are used, then there is the possibility of data being subject to recall bias. This is where participants misrecall events from the past; giving incorrect data. Additionally, as data are collected at one time point, this means that the outcome and exposure data are collected at the same time. This has the disadvantage of it being unclear which came about first. For example, questions on cancer status are asked at the same time as smoking status so it is unclear which came first and to what degree they affected one another.

  Cross sectional studies can also be prevalence studies which aim to discover the prevalence of a disease, condition or event in a sample of representative participants. If inferences are to be made from prevalence studies, then it is important that the sample is representative of the relevant population.

  The data collected could be descriptive or analytical. Descriptive studies look at characteristics, opinions and/or continuous measurements (such as blood pressure) to give a picture of the participants in the context that they are included in the study. Analytical studies are used to elicit associations between factors and/or identify risk factors for a given disease, event or condition. Analytical cross sectional studies utilise descriptive statistics before moving onto further inferential analysis.

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Research Methodologies of School Psychology

  Critical Skills

  • Ryan J. Kettler(Author)
  • 2019(Publication Date)
  • Routledge

    (Publisher)

  Cross-sectional surveys and longitudinal surveys are the two design types defined by the frequency with which the surveys are administered. Cross-Sectional Surveys Cross-sectional surveys provide a snapshot in time, with data collected at one moment from selected individuals. For example, a researcher may survey elementary, middle, and high school principals regarding a new program at a given moment in time, with the sample including both novice and experienced administrators. Such a design would provide data relatively quickly to describe each group’s views, attitudes, behaviors, demographics, and experiences (e.g., novice and experienced educators’ implementation of support behaviors or attitudes toward the program). Lane, Carter, Jenkins, Magill, and Germer (2015) conducted a statewide survey of 365 site-level administrators. The purpose of the survey study was to inform professional development in a state committed to designing, installing, and evaluating comprehensive, integrated, three-tiered (Ci3T) models of prevention to meet students’ academic, behavioral, and social needs (www.ci3t.org). In this study, researchers employed a cross-sectional design in which the primary investigators randomly selected half of the administrators in one state to participate in a survey of their knowledge and use of components constituting Ci3T models of prevention. Following approvals, half of the 1,777 public schools listed in the state Department of Education database were invited to participate. One district declined to participate, resulting in a total of 876 administrators being extended the opportunity to complete the survey. The response rate was just under 42%. This survey yielded information at one point in time. The survey did not allow for monitoring how these views shifted over time

  Sign up to read

  Learn more about book
• 

  eBook - ePub

  Psychology Around Us

  • Ronald Comer, Nancy Ogden, Michael Boyes, Elizabeth Gould(Authors)
  • 2017(Publication Date)
  • Wiley

    (Publisher)

  Truly random selection can be elusive. The members of your population who do not play video games include, for example, 3-year-olds, who probably are not interested in gaming (yet!) and probably are not capable of making the same kinds of choices about aggressive behaviour that 18- or 25-year-old persons might make. Thus, you may decide to narrow your sample to include adolescents or young adults only. Of course, such a choice would mean that your findings will be relevant to adolescents or young adults only, rather than to adults or to the entire human population. Researchers in psychology often try to choose samples that make their results relevant to the broadest possible segments of their populations of interest.

  Pick a Research Method

  Researchers have several options when designing studies to test their hypotheses (McGrath, 2011). Research methods differ in their goals, samples, and the ability of researchers to generalize their results (suggest that they might apply) to a population. Most of the methods we describe next, including case studies, naturalistic observation, and surveys, are known as descriptive research methods . They allow researchers to pursue the goal of description: to determine the existence (and sometimes the strength) of a relationship between the variables of interest. In addition to such descriptive methods, we will also describe experiments, which allow researchers to explain the causes of behaviour (see

  Figure 2.3

  ). Understanding the differences between descriptive, also called correlational, studies and experimental studies, and particularly understanding what each can or cannot say about the causes of the behaviour each observes, is critical to properly understanding psychological research.

  FIGURE 2.3 Descriptive versus experimental research Because descriptive methods and experimental methods each serve particular purposes and have different advantages and disadvantages, psychological research includes both kinds of approaches.

  Adapted with permission of John Wiley & Sons, Inc., from Carpenter, S., & Huffman, K. (2012). Visualizing Psychology , Third Edition. Hoboken, NJ: Wiley, p. 15.

  Case Studies

  A case study focuses on a single person. Medical and psychological practitioners who treat people with problems often conduct case studies to help determine whether therapeutic interventions affect their client’s symptoms (Lee, Mishna, & Brennenstuhl, 2010). A case study can be a good resource for developing early ideas about phenomena. One disadvantage of a case study, however, is that it can be affected greatly by researcher bias

  Sign up to read

  Learn more about book