Survey: A series of questions, often a standardized instrument, assessing responses to one or more topics.
Where Qualitative Research Takes Place
The chapter started by noting that qualitative approaches to research are often exploratory and seek to understand a particular phenomenon in depth. This research can take place from the lab to the field, but more often, it occurs in the field and in particular within the communities where the research question arose. An example mentioned early in the chapter was exploring reactions to building high-rise, high-density housing near a neighborhood park. In this situation, the researcher likely would reach out to people who live in the vicinity of the neighborhood park, for example, by contacting neighborhood schools or community organizations. Chapter 8 provides more information about the threats to internal validity that occur even in qualitative research.
Summary
This chapter has provided a general overview of research design as well as of internal and external validity and the components of each. You know the distinction between research where relationships are assessed and research where causal inferences can be drawn. The distinction has been made between IVs and DVs and between true IVs and quasi-IVs. We have seen how a study that started out with a correlational approach can be transformed into one that takes an experimental (causal) approach. In thinking about the number of participants needed for research, you should now be familiar with the concepts of power and effect size, and you should be able to explain the difference between Type I and Type II errors. We have reviewed the classic list of threats to internal validity from Campbell and Stanley (1963) and have expanded that list to include demand characteristics, the beliefs/attitudes of participants, the effectiveness of the cover story, and the effectiveness of manipulation checks. Finally, we have considered the different kinds of locations for research, from the laboratory to the field.
In later chapters, we will examine each of the different research designs (and associated statistics) in more depth.
If you have not had time to consider them earlier, here is the list of Revisit and Respond questions from this chapter.
3.1Explain how a quasi-IV differs from an IV and the difference in language appropriate in a correlational versus an experimental study.Identify the IV (manipulated variable) and the DV (effect or outcome variable) in the following description:“The campers were randomly assigned to two different cabins right next to each other, A and B; the camp director wanted to test the effectiveness of different kinds of mosquito netting (MN) and placed MN with white nylon fabric in Cabin A and MN with black nylon fabric in Cabin B. There was a significant difference in the number of bug bites reported by the campers in the two different cabins.”
3.2In your own words, explain the difference between Type I and Type II errors.
3.3Explain why having sufficient power is important for a study.What are the four factors from Howell (2013) that affect the power of a research design?What is the most common way to increase power?In your own words, describe what an effect size is. How is power related to effect size?
3.4Of the threats to internal validity listed by Campbell and Stanley (1963), where do you have the most control? The least control?Give an example of a demand characteristic.What are the different role attitudes participants can have, according to Adair (1973)?What is the difference between a cover story and a manipulation check?
3.5Explain the difference between external validity and ecological validity.
Practice Quiz
1 External validity measures the extent to which your design can evaluate your hypotheses.TrueFalse
2 A Type I error is also known as a false negative.TrueFalse
3 Which type of design is less likely to be used in exploratory research?Case studyExperimentFocus groupInterviews
4 Which of the following is not true of an experiment?We manipulate the independent variable.We attempt to control extraneous variables.We can infer causality.We are always able to control or measure all variables that could provide an alternative explanation of the results.
5 Random assignment means that every member of the population has an equal chance of being selected to participate in your study.TrueFalse
6 In a within-subjects design, participants are exposed to more than one level of the treatment.TrueFalse
7 Which design would you choose if the effects of one condition are likely to influence the effects of another condition and you are worried about carryover?Between-subjects designCorrelational designQuasi-experimental designWithin-subjects design
8 A Bonferroni adjustment is used to address potential problems with Type II errors.TrueFalse
9 A ________ is the examination of variables in a natural setting without manipulation.field experimentfield studylaboratory studyvirtual study
10 Most social science research uses an alpha level of .05.TrueFalse
11 Which of the following is not a threat to internal validity?HistoryMaturationInstrumentationAdequate power
12 What is the term for the lifelikeness of a research finding?Ecological validityExternal validityInternal validitySubject roles
13 If you make your test too easy, you run the risk of which of the following?Basement effectCeiling effectFloor effectHistory effect
14 Which threat to internal validity occurs when participants are selected based on extreme scores?Statistical regressionDifferential selectionExperimental mortalitySelection–maturation interaction
15 When your best friend takes part in your study and tries to answer the way she thinks you want her to, she is demonstrating a subject role.TrueFalse
16 Which word would not be used to describe a correlational result?AssociationRelationshipInfluenceCorrespondence
Build Your Skills
Take a correlational research question (e.g., the relationship between preference for the natural environment and mood), and show how you could investigate this as a quasi-experimental study and then as a true experiment.
Descriptions of Images and Figures
Back to Figure
All rooms consist of a low shelf with books next to the psychotherapist’s chair, a potted plant in the corner, and a lamp on the other side of the chair.
Western Low: A generic painting is displayed over the chair.
Western High: In addition to a generic painting displayed over the chair, the painting “Girl with a Pearl Earring,” is displayed over the shelf. The top of the shelf consists of art pieces of the western culture.
Multicultural Low: A quilt is on the wall above the chair.
Multicultural High: In addition to a quilt above the chair, a drawing with repeated motif is over the shelf. The top of the shelf consists of art pieces of different cultures.
Back to Figure
Text reads, “Correlational Design: We might predict a positive relationship between: Number of magazine subscriptions at home and fourth grade reading scores.” The accompanying illustration shows that more subscriptions leads to higher scores. Text for the illustration reads, “Relationship but no causality.”
Back to Figure
Single-Blind Experiment: Participants unaware of the condition to which they have been assigned.
Double-Blind