Psych 311 Unit 5 Part 2 study guide

Helpfulness: 0
Set Details Share
created 4 years ago by wouhib
101 views
show moreless
Page to share:
Embed this setcancel
COPY
code changes based on your size selection
Size:
X
Show:
1

Internal validity

When a research study can produce a single, unambiguous explanation for the relationship between two variables

Threat: Any factor that allows for an alternative explanation for the results (extraneous and confounding variables)

2

Extraneous variables

  • Any variable in a research study other than the specific variables being studied
  • For example, individual differences, personality variables, environmental variables
3

Confounding variables

  • Extraneous variables (usually unmonitored) that change systematically along with the variables being studied
  • Confounding variables provide alternative explanations for results and, therefore, are threats to validity
  • Types: Environmental Variables, individual differences, time-related variables
4

Environmental variables

  • Examples: time of day, size of room, gender/race/ethnicity of experimenter
  • If the two treatments are administered in different environments, then internal validity is threatened
  • Threat to ALL study designs
5

Individual differences

  • Assignment Bias: When the process used to assign different participants to different treatments produces groups with noticeably different characteristics
  • Results could be explained by treatment… or by individual differences
  • Threat to study designs that compare differences between groups
6

Time-related variables

  • Includes: history, maturation, instrumentation, testing effects, regression toward the mean
  • Threat when studying a variable over time
7

Specific threats to internal validity

History, maturation, instrumentation, testing effects and statistical regression

8

History

  • When outside events might affect some participants more than others
  • Differences can be explained by outside events
9

Maturation

  • When participants might experience physiological or psychological changes
  • Differences in treatment can be explained by changes
10

Instrumentation

  • Also called instrumental bias or instrumental decay
  • When the measurement instrument changes from one treatment to another
  • Differences may be explained by changes in instrument
11

Testing effects

  • Testing Effects –possible change due to general participation in a previous treatment
  • Carry-Over Effects – possible change due to experiencing a specific treatment
  • Both Testing Effects and Carry-Over Effects can be considered Order Effects
  • Change in performance may also be caused by fatigue
12

Statistical regression

  • Also called regression toward the mean
  • If participants are selected based on extreme scores (high or low), validity may be threatened
  • Extreme scores have a tendency to move toward the mean (regress) when measurements are repeated
  • A change toward more average scores could be explained by statistical regression
13

Balancing internal validity with external

card image
  • Trade offs necessarily occur, when trying to increase one type of validity
  • Different strategies have different goals, and also different levels of internal and external validity
14

Balancing external validity with internal

card image
15

Research strategies and validity

  • Descriptive, correlational, nonexperimental prioritize real world settings and, therefore, have stronger external validity
  • Experimental research is highly controlled – and has strong internal validity
  • Quasi-experimental studies are often in the middle, attempting to mimic experiments while taking place in real-world settings
16

Artifacts

card image
  • External factors that may influence or distort measurements
  • Can threaten both internal and external validity
17

Experimenter bias

  • When study findings are influenced by an experimenter’s expectations or beliefs regarding the outcome of the study
  • Threat to External Validity: results may be specific to research
  • Treat to Internal Validity: data may show treatment effect but was influenced by experimenter
18

Demand characteristics

  • Cues that suggest the hypothesis to participants and cause them to react a certain way
  • Threats to internal & external validity: Greater in lab studies, but still possible in field studies
19

Participant reactivity

  • When participants alter behavior because they are being studied
  • Threats to internal & external validity: Lab studies have greater likelihood for reactivity than field studies
20

Exaggerated variables

  • Most research undertaken in order to show relationships between variables
  • Differences in treatment conditions can be exaggerated to increase the chance that scores in one treatment are noticeably different from those in another (this can be very useful!)
  • Threats to internal & external validity: The more exaggerated the variables, the less likely they are to translate to real life
21

Alternative explanation

  • The possibility that a confounding variable rather than the independent variable of interest caused changes in the dependent measure represents