Exam 1 - Testing and Evaluation

Question 1

Degree to which conclusions about causal relationships can be made (e.g. cause and effect)
—Did the treatments cause the outcome to occur, or did other extraneous factors cause the outcome?
—Well-controlled testing procedures and conditions ensure internal validity

Answer 1

Internal Validity

Question 2

Degree to which the (internally valid) results can be generalized to other populations, settings, or experimental treatments.
—If it works for Arnold, will it work for similarly built individuals? Will it work for everyone?
—Could the well-controlled findings of a study using prison inmates be applied to the general population?
——–Good internal validity, bad external validity

Answer 2

External Validity

Question 3

• Learning effect
• subject selection bias
• bias in regards to the specific variable beingg measured
• reactive effects of experimental setting (would the same thing happen if it occurred in a natural setting)
• multiple-treatment interference (effects of earlier treatment are irreversible)

Answer 3

Threats to external validity

Question 4

• learning effect vs training effect
• instrument error
• investigator error
• Passage of time
• Testing instructions
• Encouragement and feedback
• Similarity of Test-Retest conditions (social, environmental, procedures, mental, physical, nutritional)

Answer 4

Threats to Internal Validity(For the individual)

Question 5

• Experimental mortality (drop-outs)
• Non-homogenous groups (ages, gender, training status, medical history)
• Statistical regression (returning to group mean), when groups determined from extreme scores
• Biased group selection (intentionally or unintentionally putting all the best athletes in one group

Answer 5

Threats to Internal Validity(For the groups)

Question 6

• Repeatability of a test conducted by the same observers
• Necessary for determining the influence of a training program or supplement for causing a change

Answer 6

Intrarater (Test-Retest) Reliability

Question 7

• degree of agreement between different raters/observers to give consistent estimates of the same phenomenon
• Necessary to compare your results with those of others

Answer 7

Interrater (Inter-Observer) Reliability

Question 8

Systematic Error
–Constant error affects all scores equally (e.g. learning effect)
–Bias affects certain scores differently

Random Error (e.g. imprecision)
–Affected by luck, alertness, attentiveness by tester, and normal biological variability

Total Error = Systematic + Random

Answer 8

Threats to reliability