Chapter 7: Nonexperimental Research PDF

Chapter 7: Nonexperimental Research Overview of Nonexperimental Research Nonexperimental research lacks: 1) 2) 3) As a result, nonexperimental research cannot provide strong evidences that the IV causes the change in the DV. Nonexperimental research can serve an important purpose when it is not f...

Chapter 7: Nonexperimental Research Overview of Nonexperimental Research Nonexperimental research lacks: 1) 2) 3) As a result, nonexperimental research cannot provide strong evidences that the IV causes the change in the DV. Nonexperimental research can serve an important purpose when it is not feasible or ethical to manipulate the IV or randomize participants to different conditions Overview of Nonexperimental Research Some useful applications of nonexperimental research: 1) Single IV with one condition e.g. 2) Investigation of noncausal relationships e.g. 3) IV cannot be manipulated or participants cannot be randomly assigned to different conditions e.g. 4)Exploratory research e.g. Types of Nonexperimental Research 1) Single variable research: Example 1: One group, pretest-posttest: A new fourth grade math curriculum is introduced and students' math achievement is assessed in the fall and spring of the school year. Improved scores on the assessment are attributed to the curriculum. Weakness of this design? Types of Nonexperimental Research Example 2: Measuring the self-esteem of children who have been bullied Weakness of this design? Types of Nonexperimental Research 2) Correlational research: focus is on the statistical relationship between two variables; no manipulation, no random assignment Example: 3) Quasi-experimental research: IV is manipulated, but participants are not randomly assigned to conditions Example: Types of Nonexperimental Research 4) Qualitative research: To gain an understanding of the underlying reasons and motivations of participants’ experiences To uncover prevalent trends in thoughts and opinion These findings are non-numerical and are not arrived at by statistical or quantitative procedures Examples: How do patients perceive doctors in a hospital setting? What are young women’s experiences on dating sites? What are non-experts’ ideas and opinions about climate change? Types of Nonexperimental Research Examples of methods used in qualitative research: Thematic analysis: applied to a set of texts (e.g. interview transcripts) The researcher closely examines the data to identify common themes – topics, ideas and patterns of meaning that come up repeatedly. Example of thematic analysis Example of thematic analysis Comparison of internal validity of various research designs What kind of research design? And why? 1) A researcher conducts detailed interviews with unmarried teenage fathers to learn about how they feel and what they think about their role as fathers and summarizes their feelings in a written narrative. 2) A researcher measures the impulsivity of a large sample of drivers and looks at the statistical relationship between this variable and the number of traffic tickets the drivers have received. What kind of research design? And why? 3) A researcher randomly assigns patients with low back pain either to a treatment involving hypnosis or to a treatment involving exercise. She then measures their level of low back pain after 3 months. 4) A college instructor gives weekly quizzes to students in one section of his course but no weekly quizzes to students in another section to see whether this has an effect on their test performance. 1) Correlational Research A type of nonexperimental research design where the association between variables are assessed, with no effort to control extraneous variable Purpose: 1) There is no causal relationship assumed to underlie the observations e.g. 2) The relationship is causal, but it is unfeasible and/or unethical to manipulate the IV  1) Correlational Research Other features: The terms IV and DV do not apply to correlational research designs Does not have to involve quantitative variables Correlational or Experimental?  Data collection in correlational research A) Naturalistic observation: observing people’s behaviour in an environment in which it typically occurs e.g. field research Researchers should make their observations as unobtrusively as possible Avoid the Hawthorne effect Make people habituated to your presence when distancing your proximity is not an option Becoming an insider or part of the group, becoming familiar Data collection in correlational research Issues to deal with in naturalistic observation: 1) Sampling: which participants will be observed and under what conditions? 1) Good to have participant inclusion criteria 2) Measurement: which behaviours will be observed and how? Coding: requires clearly defining a set of target behaviours Data collection in correlational research B) Archival data: data that have already been collected for some other purpose Content analysis: systematic approach of identifying keywords, phrases, or ideas and documenting all occurrences of them; -occurrences can be analyzed in a variety of ways -frequency counts, ratings, Example: 2. Quasi-experimental Research “quasi” = “resembling” When group’s can’t be randomly assigned Pro: Direction of causation can be determined because IV is manipulated Con: Confounding variables cannot be eliminated due to lack or random assignment (i.e. other differences between conditions could be attributed to group differences 2. Quasi-experimental Research a) Non-equivalent group design: groups are dissimilar in some ways due to lack of random assignment -because random assignment isn’t possible; e.g. groups who are neurotic vs groups who are not in personality b) Pretest-posttest design: DV is measured prior to and after manipulation e.g. Measurement of student attitudes prior to and after anti-drug program Problem? 2. Quasi-experimental Research Alternative explanations for changes observed in pretest-posttest design a) History: b) Maturation: changes due to growth and learning Having a control group in wait can help with this. c) Regression toward the mean: individuals who demonstrate extreme performance on one occasion will tend to score less extremely in a subsequent occasion Example: people with extreme depression scores  treatment  their scores become more average during the post-test Was it because of the intervention, or was it just because people tend to not score as extreme the 2nd time as the 1st time? 2. Quasi-experimental Research d) Spontaneous remission: tendency for medical or psychological problems to improve over time without any form of treatment. Example: Waitlist group in treatment of depression 2. Quasi-experimental Research c) Interrupted time series design: Advantage: allows you to track the Trends and development over a time period. Useful for analyzing the data Longitudinally 2. Quasi-experimental Research d) Combination design: 1. Treatment group: Pretest – Treatment – Posttest 2. Control group: Pretest – No treatment – Posttest Advantage: the control group allows you to interpret your treatment effects more accurately