Module 2 chpt 9, 10, 11

Question 1

Validity

Answer 1

soundness of study evidence - unbias as well as grounded

is the method really measuring what your concept is (is it depression or loneliness)

Question 2

reliability

Answer 2

accuracy and consistency of information obtained in the study

Question 3

credibility

Answer 3

aspect of trustworthiness - research methods inspire confidence and results and interpretations are truthful

Question 4

bias

Answer 4

influence that produces a distortion or error.

• participants lack of candor
• researcher subjectivity
• sample imbalance
• faulty data collection
• inadequate study design
• flawed implimentation

Question 5

random bias

Answer 5

“noise: in the data - bias in one direction or another

Question 6

systematic bias

Answer 6

consistent and is in one direction ie a scale weighs everyone 2 lbs heavier

Question 7

research control

Answer 7

attempts to eliminate contaminating factors that might obscure the relationship btw variables

Question 8

confounding variables

Answer 8

contaminating factors - must be held constant to reduce influence

Question 9

randomness

Answer 9

having features of the study established by chance rather than by researcher preference
Controls confounding variables and reduces bias

Question 10

Randomized control trial

Answer 10

experimental research - researchers are active angents
GOLD STANDARD
MOST convincing if one variable has a causal relationship with another
Manipulation, control, and randomization

Question 11

validity

Answer 11

property of an interference

Question 12

threats to validity

Answer 12

introduce features to minimize potential threats - strengths validity

Question 13

internal validity

Answer 13

it is the IV rather than something else that caused the outcome

Question 14

external validity

Answer 14

inferences about observed relationships will hold over variations in persons in the real world setting

Question 15

controlling confounders

Answer 15

randomization - most effective method
crosssover - participants serve as their own controls - can cause time influence (intervention A caused change to intervention B)
homogeneity - pick for homogeny for the cofounders - does not translate to those not in the research group
stratification/blocking - one variable against another ie gender (males in one black, females in another)
matching - using characteristics to create comparable groups - must know cofounders in advance
statistical control - ANCOVA statistical removal of confounding variable

Question 16

statistical power

Answer 16

used to detect exiting relationship among variables

larger sample = larger statistical power