Research Design

1
Research Design

• Understanding the basic premises that influence
  your capstone project

2
Types of Research

• Basic - predictions based on existing theory
  comparison of findings to those predictions

Importance of the weight of all evidence that
is, consistent with or not consistent with
the theory.
3
Types of Research

• Applied - practical in nature solving
  immediate problems

Examples evaluating effectiveness of
supplements, finding the best training methods,
identifying those at risk
4
Types of Research

• Cross Sectional - looking at different groups
  at the same time (efficient use of time and
  resources)

Groups are different based on certain variables
(e.g., age, fitness level, ethnicity, major).
Note. homogeneity
5
Types of Research

• Longitudinal - looking at the same group over
  time (time consuming and complicated) often a
  control group

Examples Framingham Heart Study, Harvard
Physician Study, Stanford Community Fitness Study
6
Types of Research

• Qualitative
• observation of natural situations
• naming the data
• no manipulation
• flexibility in methodology and interpretation
• researcher skills

• Quantitative
• control of situational variables
• quantifying the data
• manipulation of IVs treatment conditions
• extensive planning and laboratory setup
• tests and procedures

Descriptive
Experimental
7
What will you be doing?

• Broadening the scope
• different population
• different setting
• controlling previously uncontrolled variables
• Replication
• resolving inconsistencies
• improving methodologies

8
The Variable...
a characteristic (attribute or dimension) that
will exhibit different values under different
conditions
Research Goal to understand how variables behave
or react to different conditions
9
Types of Variables

• Independent variables manipulated by
  investigator (selected / levels decided on)
• Dependent variables measured by the investigator
  (outcome)
• Control variables held constant during
  experiment (delimiters)
• Confounding variables unintentionally allowed to
  vary during the experiment (limitations)

10
Variance of the dependent variable(s) will be
either...

• Primary expected or desired related (at least
  in part) to the effect of the independent
  variable
• Secondary consistent, but unwanted related to
  measurement methods and procedures
• Error inconsistent and unwanted related to
  imprecise measurement (often for unknown reasons)

11
Research Validity

• Internal - certainty with which the results can
  be attributed to the independent rather than
  confounding variables
• Threats to Internal Validity
• local history
• maturation of subjects
• pretesting / learning effect
• instrumentation
• statistical regression
• selection of subjects
• mortality

12
Maximizing Internal Validity

• Random selection / assignment
• Use of control groups
• Careful measurement
• Appropriate analysis and interpretation

13
Research Validity

• External - generalizability of the results to
  other people and / or other situations
• Threats to External Validity
• subjects
• pretest sensitization
• expectancy
• Hawthorne Effect
• overgeneralizing

14
Maximizing External Validity

• Detailed definition of study population
• Proper sampling procedures
• Replication (similar results under range of
  conditions)
• Establishing internal validity