CHAPTER 2: Conducting Psychological Research1

Chapter 2: Conducting Psychological Research

A.LEARNING OUTCOMES. After studying this chapter students should be able to:

  • Discuss sources for getting ideas about a research topic.
  • Identify major online psychological databases.
  • Describe ways to broaden and narrow a literature search.
  • Describe the main sections of a research article and different types of review articles.
  • Explain how hypotheses are formed, both inductively and deductively.
  • Describe characteristics of a good hypothesis.
  • Describe general types of approaches to conducting research.
  • Discuss the concepts of internal and external validity in relation to laboratory and field research settings.
  • Identify some key issues in preparing to conduct a study.
  • Distinguish between qualitative and quantitative data analysis.
  • Describe the general uses of descriptive and inferential statistics, and identify measures of central tendency and dispersion.
  • Explain the concept of “statistical significance” and provide two general types of errors that can occur when researchers make inferences from data.
  • Describe diverse ways in which researchers communicate their findings to the scientific community.
  • Describe some benefits of building theories, and give the characteristics of a good theory.
  • Discuss whether the results of a study should be viewed as proving or disproving a theory.

B.KEYWORDS

Confounding variable / Field study / Operational definition
Content analysis / Hypothesis / Population
Cross-sectional research / Independent variable / Qualitative research
Deductive reasoning / Inductive reasoning / Quantitative research
Dependent variable / Inferential statistics / Range
Descriptive research / Internal validity / Sample
Descriptive statistics / Law of parsimony / Sequential research design
Effect size / Longitudinal research / Standard deviation
Evidence-based treatments / Mean / Statistically significant
Experiment / Median / Theory
External validity / Meta-analysis / Variance
Field experiment / Mode

C.BRIEF CHAPTER OUTLINE

I.Initial Observations and Questions

  1. Personal Experience and Daily Events
  2. Prior Research and Theory
  3. Real-World Problems
  4. Serendipity

II.Gathering Background Information

  1. Searching Scientific Databases
  2. Obtaining Articles
  3. Reading Research Articles
  4. Reading Review Articles

III.Forming a Hypothesis

  1. Forming Hypotheses Inductively
  2. Forming Hypotheses Deductively
  3. Characteristics of a Good Hypothesis

IV.Designing and Conducting a Study

  1. Approaches to Conducting Research
  2. Planning and Performing the Study
  3. The Role of Sampling

V.Analyzing Data and Drawing Conclusions

  1. Quantitative and Qualitative Analysis
  2. Descriptive and Inferential Statistics
  3. Drawing Conclusions

VI.Reporting the Findings

VII.Building Knowledge and Theories

  1. Characteristics of a Good Theory
  2. Proof and Disproof

D. EXTENDED CHAPTER OUTLINE

*Much of this summary is taken verbatim from the text.

Introduction

This chapter is about the research process, beginning with how one identifies a research topic to how one uses data collected from research to build theories and knowledge.

Part I: Initial Observations and Questions

The ideas for research studies come from a variety of sources. This section describes how personal experience and daily events, prior research and theory, real world problems, and serendipityare used to develop testable research questions.

  1. A person can generate a number of good research questions simply by reflecting on his or her personal experience and daily events. One example of a daily event that has led to an enormous amount of research is the World Trade Center tragedy of 2001. Since 9/11 researchers have examined the extent to which the attacks have affected Americans psychologically, how future attacks might be prevented, and what factors are necessary to ensure adequate responses to disasters. While this is an extreme daily event, it does go to show that an event, major or otherwise, can be used to generate important research questions.
  2. A good number of research topics stem from prior research and theory. Most good empirical papers will leave the reader asking questions about what might be done next. For example, if a paper describes the frequency that teenagers lie, a subsequent question might ask whether the frequency of lying is different for boys and girls.
  3. Real-world problems are also used to generate research questions. Noticing that some children develop PTSD following a severe stressor while others do not may lead to an investigation of the factors that can contribute to psychological resilience.
  4. Serendipity is the accidental discovery of some important thing. Many scientific discoveries (i.e., penicillin, rubber) have been made through serendipity.

Part II: Gathering Background Information

This section describes a number of ways in which scientists acquire background information relating to particular research topics. Students will learn that a review of the literature is an important step in the research process.

  1. Most literature reviews are done by searching scientific databases, and good literature reviews take time, primarily because the person must refine his or her search parameters to ensure that the literature identified is relevant to the specific research question. There are many scientific databases, including:
  2. PsycINFO and PsycARTICLES are APA databases that collectively track thousands of journals.
  3. Some discipline-related databases such as MEDLINE (medical research) overlap with psychology and can also be used to investigate research questions in psychology.
  4. General search engines, including Google Scholar, are also helpful in conducting a literature review. However, the results obtained from general search engines are often very broad and generate literature references unrelated to the specific topic.
  5. For all databases it is important for the user to understand how to both narrow and broaden a search. It’s kind of like the “Three Bears” story—you don’t want too much literature or too little literature. You want an amount that’s just right.
  1. Once you have identified the relevant literature, the next step is to actually obtain the articles. Today many articles are available electronically in PDF form; the database you’re using will tell you if the article you want is available and whether there is a cost to download it. Articles can also be found in print; if your library does not have the article (either in print or in electronic form) you can ask the librarian to obtain it via interlibrary loan.
  2. Research articles, those that describe an empirical study, usually have a similar structure. Typically they are preceded by an abstract, followed by introduction, method, results, discussion, and reference sections.
  3. Review articles are different from research articles in that they are not based on an original research project. Instead, review articles are written by experts who integrate the literature on a particular subject and summarize it in a qualitative or quantitative way.

Part III: Forming a Hypothesis

Students should recall from Chapter1 that a hypothesis is a testable statement that is tentatively held to be true. “If–then” statements are often used to form a prediction between the independent and dependent variables. Still, the prediction must be based on logic and not simply on a hunch. Two forms of reasoning may be used to form a hypothesis: inductive reasoning and deductive reasoning.

  1. Inductive reasoning is the process by which several specific facts are used to form a more general statement.
  2. Whether a hypothesis is generated via inductive or deductive reasoning, all good hypothesesmust be testableand falsifiable. That is, there must be observable and measureable datato either support or refute the hypothesis.

Part IV: Designing and Conducting a Study

  1. There are several approaches to conducting research. Research may be:
  2. Quantitative or qualitative. Quantitative research relies on numbers and numerical analyses to describe behavior, whereas qualitative researchis more concerned with capturing the essence of behavior in a nonnumerical way. In qualitative research information from many people are examined to determine whether underlying themes and patterns are common among them. These consistencies are then described in a purely verbal way.
  3. Experimental or descriptive (nonexperimental). In experimental research a scientist seeks to establish a causal relationship between an independent variable (IV)and a dependent variable (DV). Inferring causality is due to the researcher directly manipulating an independent variable and later measuring a dependent variable. Descriptive research also seeks to determine whether variables are related, but the methodology does not include manipulation of an independent variable. As such, causal statements about the IV–DV relationship cannot be made. Because of the inability to make causal statements, descriptive research is also often referred to as non-experimental research.
  1. Conducted in the laboratory or in thefield. Laboratory research is highly controlled. This level of control enhances internal validity, which is the degree to which we can be confident that our study truly demonstrated that one variable had a causal effect on another. It is usually difficult, however, to make laboratory research simulate real life. That said, when a scientist wants to study behavior in a more natural state he or she conducts field research. The “field” is an environment as it naturally exists. Doing field research increases the external validity, or generalizability, of the study to real-world situations. Research need not sacrifice one type of validity for another. It is possible to conduct an experiment (high internal validity) in the field (high external validity).
  2. Be designed to study age and developmental differences. Cross-sectional research examines people of different ages at the same point in time. In contrast, longitudinal research examines a single group of people across different points in time. The two methods can also be combined to create a sequential design. In sequential designs many cohorts are examined over time. While they are more elaborate and costly, sequential designs encompass the advantages inherent to the cross-sectional and longitudinal designs.
  1. When it comes to planning and performing the study one must consider:
  2. ethics
  3. operationism
  4. which design will best answer the research question
  5. who will comprise the sample
  6. who will conduct the research.
  7. Therole of sampling in research is particularly important. A sample is a subset of a population, and the way in which a sample is selected and measured has a huge impact on how well the data represent the population.

Part V: Analyzing Data and Drawing Conclusions

  1. Once data are collected the researcher must analyze it. Quantitative analysis involves the use of statistics to interpret behavior, whereas qualitative analysis is nonmathematical and uses verbal descriptions to illustrate behavioral patterns and themes. Often, researchers will utilize both types of analysis in their data analysis.
  2. There are two general types of quantitative analysis.
  3. Descriptivestatistics are used to organize and summarize data and include a measure of central tendency (mean, median, and mode) as well as a measure of dispersion (range, variance, standard deviation).
  4. Inferential statistics are used to draw conclusions about a population based on sample data. In an experiment, a researcher examines whether an independent variable (i.e., Xanax) affects behavior (i.e., anxiety). After participants take the drug, or a placebo, anxiety is measured. Inferential tests determine whether the mean anxiety score is due to whether participants took Xanax or something else (sampling error). If the difference in anxiety is found to be due to the drug, the difference between the group means is said to be statistically significant.
  5. After data are analyzed the experimenter must draw conclusions about the relationship between the variables that were under examination. Although inferential statistical tests provide a level of confidence about our conclusions, they are not ironclad and mistakes can still be made (Type I and II errors will be discussed in Chapter 9).

Part VI: Reporting the Findings

There are many ways to communicate research findings. The most esteemed way to disseminate research is to publish it in peer-reviewed professional journals. Books, book chapters, and poster presentations are other ways to communicate your work. The American Psychological Association (APA) provides guidelines on how to create research reports in various media.

Part VII: Building Knowledge and Theories

As research accumulates, scientists may form a theory. Theories function to organize knowledge into a coherent whole, to help understand and make predictions about new situations for which there currently is no empirical data, and to generate interest on a topic and provide a focus for new research.

  1. Characteristics of a Good Theory
  2. Testability and specificity
  3. Internal consistency and clarity
  4. Empirical support
  5. Parsimony
  6. Scientific impact
  7. Proof and Disproof. Evidence to support a theory does not absolutely prove it to be true. According to Baumeister (2008) the entire enterprise of psychological theorizing and research is more probabilistic than absolute.

C.LECTURE AND CLASSROOM ENHANCEMENTS

PART I: The Initial Observations and Questions

In this section students will learn how research topics can come from a variety of sources. Students will learn that they can look to their own personal experience, as well as current psychology literature, to identify good research ideas.

A.Lecture/Discussion Topics

  • How do everyday events lead to research ideas? In the 1960s a young woman named Kitty Genovese was brutally stabbed and murdered outside of her apartment building. It was later found that many of her neighbors heard Kitty cry for help but no one actually went to her rescue. This event sparked the curiosity of Bibb Latané and John Darley who, in 1969, published a paper describing what is now commonly referred to as the Bystander Effect. A plethora of research has been done since then to support the idea that as the number of people in the area increase, the likelihood that any one person will take action is significantly reduced. This story is an excellent example of how personal experience and daily events can spark important and fruitful research topics.

B.Classroom Exercise

  • Establishing a research project. Many experimental psychology classes require students to conduct a research project. Based on my experience, students tend to have difficulty coming up with a topic to investigate. To help them get started have them write down at leasttwo topics that spark their curiosity, based on (1) personal experience, (2) what they’ve learned in other psychology courses/psychology research, and (3) real-world problems. Next, have students search for literature on one of the topics using your library’s database resources. Have them keep a record of the keywords and phrases they used to conduct their search, as well as how much (or how little) information the searches generated. This record can then be used to illustrate how the words AND, NOT, and OR can help refine literature searches.

C.Web Resources

  • This link describes ten serendipitous discoveries that revolutionized the world. (#1 on the list describes how a dirty workstation led to the discovery of penicillin!) Although serendipity is typically very happenstance and nonscientific by its very nature, students will understand that it has definite value in research and can lead to the empirical study of particular things.
  • This link summarizes the Kitty Genovese story and the bystander effect as described by Latané and Bibb.

D.Additional References

  • Eighteen serendipitous discoveries in psychopharmacology:

Baumeister, A.A., Hawkins, M.F. & Lopez-Munoz, F. (2010). Toward standardized usage of the word serendipity in the historiography of psychopharmacology. Journal of the History of the Neurosciences, 19, 253–270.

PART II: Gathering Background Information

In this section students will learn how to acquire a body of literature related to a specific research topic. Students will also learn how to read empirical and review papers.

A.Lecture/Discussion Topics

  • In many of my classes students are required to summarize and discuss a peer-reviewed empirical research article. Each semester at least one student completes the assignment based on a review article. Clarify the difference between empirical research articles and review articles. Describe how empirical articles can be easily identified because they always include methods, results, and discussion section whereas review articles do not. Let students know that some search engines, including PsycINFO, enable the user to check a box requesting peer-reviewed literature.

B.Classroom Exercise

  • See exercise in Part I.
  • Prior to class search online for a particular topic (i.e., Stress AND neurogenesis using a professional database). Print the first page of the results and hand it out to your students. Have them try to identify whether the paper is a review article or an empirical article and to provide rationale for their answer. You may want to follow up this exercise with a trip to the library for them to inspect the full article. While the title is a good index of the nature of the paper, it may still be difficult to make a clear distinction.

C.Web Resource

  • This link is for a PDF file of a PsycINFO Quick Reference Guide that is sure to help students conduct a better literature search.
  • This link is to a PowerPoint presentation on how to effectively use PsycINFO.

D.Additional References

  • Conducting good literature reviews:

Fink, A. (2009). Conducting research literature reviews: From the Internet to paper. SAGE Publications, Incorporated.

Cooper, H. M. (1988). Organizing knowledge syntheses: A taxonomy of literature reviews. Knowledge in Society, 1(1), 104–126.

PART III: Forming a Hypothesis

This section describes how hypotheses may be formed through inductive and deductive logic. While the two types of reasoning are described independently, it is important to convey to the student how each works cyclically to build existing knowledge.

A.Lecture/Discussion Topics

  • Which is better: inductive or deductive logic? The scientific method is often described as a process of inductive logic. Scientists collect data, and, based on the results, they develop a more general theory. However, the deductive method of reasoning is just as important in science. In deductive logic, one takes a top-down approach such that after a general theory is stated, specific tests are applied to determine whether evidence exists to support it. Beginning psychology students tend to approach research projects from a deductive perspective. Recently, a group of my students asked the question “Does cinnamon affect memory?” This question was not based on a review of the literature, but was asked just out of sheer curiosity. To answer the question the students developed an experiment that would allow them to collect data that would either support, or not support, a relationship between cinnamon scent and memory.

B.Classroom Exercises