Methods Flashcards
What elements/parts are in the Methods Section?
Subjects/Participants
Experimental Materials
Procedure
Data Analyses/ Statistical Analyses
Significance of Method Section
-if the methodology is wrong, then all goes wrong
-should provide the research design to answer the research questions (or hypothesis)
-critical for evaluation of the results and discussion sections
Sample Size
-larger is better
- if the goal is to generalize data, the sample size has to be larger
-between-subject designs require a larger n than within-subject designs
-standardized test batteries and survey research require larger n
-if small group differences are expected, larger n should be used to demonstrate statistically significant differences
Subjects Protection
Institutional Review Board (IRB), protection of subjects from risk for human subjects
Health Insurance Portability and Accountability Act, the protection of human subjects’ privacy and confidentiality
Two experimental validities related to subjects
Internal Validity- make sure you have unbiased and well-controlled subjects
- the extent to which an experiment accurately measures what it intended to measure
External Validity- make sure you have an unbiased and enough size of sample for generalization of your findings which is important in a clinical study
Threats to Internal and External Validity
-Internal
History
Maturation
Reactive Pretest
Instrumentation
Statistical regression
Differential subject selection
Attrition
Interaction of factors
-External
External Generality
Inferential
Logical
Ways to control external validity better
Subject randomization- it can help that the sample represents the population better
Larger sample size- better represents the population and improves generalizability
Experimental Materials
- authors present enough details about materials
- define IV & DC clearly and appropriately
-behavioral materials or instruments described with detailed information
-Instrumentation measurement devices and their calibration
Procedures
Tasks and protocol- sequence of tasks performed by the subjects
Test Environment- noise levels, settings, clinic or lab? how many hours per day?
Instructions- for the completion of their tasks must be clear and appropriate for the population being measured
Data collection bias- data recording errors due to dishonesty, carelessness, inconsistency, and lack of knowledge, observers interaction with the subject changes the subjects behavioral responses
Data analysis/statistical analyses
The last portion of the method section describes how data will statistically be analyzed