Nonexperimental Methods I: Descriptive Methods, Qualitative
Research, and Correlational Studies
Lecture, Chapter 4
Typically used in exploratory studies - to explore a person’s life, a behavioral phenomenon, a culture, or a person’s specific state or situation on a deep level.
Is more likely to be internally, but not externally, valid.
lends itself to inductive research, rather than deductive research.
One of the strongest criticisms of qualitative research is the subjectivity of the researcher, in that a researcher’s perspective can influence the way phenomena are observed and recorded.
Internal validity (accuracy of measurement)
Credibility (trustworthiness and authenticity)
Types of Qualitative Studies
Biography-exploring life of an individual
Phenomenology-understanding a phenomenon
Grounded theory-inductive process of theory development from field datum
Ethnography-describing a cultural or social group
Case study-in-depth analysis of a single case or one aspect of multiple cases
Qualitative Data Analysis
Coding is a method of qualitative analysis involving labeling common emerging themes from a data set. In coding, the researcher is the computer.
Read the data thoroughly and completely, making a conscious effort to notice patterns.
Identify themes that emerge from the data and label them, or “code” them.
Use the same code to label common themes.
Make a list of your codes and note the frequency of each code
Descriptive methods: used to “describe;” no manipulation of IV, therefore no causation is established.
Archival and previously recorded data – very little control over sample, extraneous variables during collection, or physical damage to the datum over time.
Choosing Behaviors and Recording Techniques
Time sampling – observing behavior at different times
Situation sampling – observing same behavior in different situations
Role of Consistency
Interobserver reliability – degree of consistency among observers’ ratings
(# agreements / # observations) X 100 = % agreement
Should have 85% agreement among observers to claim reliability.
Correlation studies examine the relationship between 2 variables.
Types of correlations:
Remember that many correlations occur due to extraneous variables. (Ex. Relationship between shoe size and candy sales. What variable needs to be controlled?)