HOC8:EXPERIMENTS

Question 1

What is an experiment

Answer 1

Used in deductive research, for cause-and-effect studies

Question 2

what is an independent variable

Answer 2

variable that researcher directly manipulates

Question 3

what is an dependent variable

Answer 3

observable outcome

Question 4

what is a treatment

Answer 4

manipulation of IV to by creating different levels of IV to assess impact on Dv

Question 5

what is a pre-test and post-test

Answer 5

measurement of DV b4 and after

Question 6

what is a subject

Answer 6

it is a unit of observation

Question 7

what does control mean

Answer 7

any means to keep other factors of a situation constant

Question 8

explain field experiments

Answer 8

studying a subject in its natural environment
- Definition: Conducted to establish cause-and-effect relationships using a natural, not contrived, environment.
- Setting: Takes place in the subjects’ natural environment where they typically function.
- Subjects: Can include employees, consumers, managers, children, etc.
- Interference: Limited interference by the researcher to maintain the authenticity of the setting.
- Purpose: Investigates real-world scenarios to draw conclusions about causal relationships.

Question 9

explain what a lab experiment is study in artificial environment

Answer 9

is study in artificial environment
- Definition: Controlled experiments conducted in a controlled and artificial environment created by the researcher.
- Purpose: To establish cause-and-effect relationships by manipulating variables.
- Characteristics: High control over variables, potential for high internal validity.
- Interference: Researcher actively creates the experimental environment.
- Examples: Psychological studies, medical trials, laboratory research.

Question 10

what are the 7 steps of conducting an experiment

Answer 10

1. select relevant variables
2. specify the treatment levels
3. control experiment environment
4. choose the experimental design
5. select and assign the participants
6. pilot test, revise and test again
7. do the experiment and analyze data

Question 11

explain the meaning of casual relationship

Answer 11

• Definition: A relationship where changes in one variable cause changes in another, establishing cause-and-effect connections.
• Controlled Environment: In organizational settings, controlling covariates is challenging, especially in natural settings where events occur spontaneously.
• Field Experiment Challenge: Even in field experiments, maintaining a completely natural environment is difficult once participants become aware of the experiment.
• Examples: Studying leadership effects on productivity, where controlling all variables is not always feasible in real-world scenarios.

Question 12

what is treatment ( or manipulation)

Answer 12

creating different levels of the IV to assess impact on DV -> is required to examine casual effect of an IV on a DV

Question 13

what does control mean

Answer 13

Control

• if we want to examine cause-and-effect relationship between X ( IV) and Y ( DV) → then other factor , A , might influence Y
• example : studying effect on training on learning effectiveness
  
  • X : training (IV)
  • Y : learning effectiveness (DV)
  • A : previous web experience
• A if a confounding ( or contaminating ) factor ⇒ may distort effect of X on Y → A must be eliminated
• How to asses true effect of training on learning effectiveness? ⇒ by controlling previous web experience
• how practically? by not including in the experiment those employees who had some web experience

Question 14

what are matching groups in controlling contamination

Answer 14

matching various groups by picking the confounding/contaminating characteristics and deliberately spreading them across groups
- Less effective. Factors not always known

Question 15

what are randomization groups in controlling contamination

Answer 15

• Definition: Randomization involves randomly assigning members to different groups, ensuring each member has an equal chance of being in any group.
• Analogy: Comparable to throwing names in a hat and drawing them randomly.
• Objective: Achieves an equal distribution of confounding variables among groups, making each group comparable.
• Importance: Minimizes the impact of potential confounding factors, enhancing the validity of experimental results.

Question 16

what is the comparison between matching groups and randomization

Answer 16

matching groups = less effective since :
-possibly confounding factors are not always known ,
-even if cofounding factors are known , then matching is not always possible
randomization =
- through normal distribution , confounding factors will be spread , and sample bias reduced
-most reliable with larger samples

Question 17

what is internal validity

Answer 17

confidence we place in cause-and-effect relationship
->Evaluates the study’s internal validity, considering the choice of independent variable (IV) and the removal of confounding factors.
->Significance: The higher the internal validity of the research, the more confidently we can assert a cause-and-effect relationship between variables A and B.

Question 18

what is external validity ( generalizability)

Answer 18

= how does It generalize across people, settings (Larger in field)
“does the observed relationship generalize across people , settings and times?” → “ to what extent are the results found in lab setting transferable or generalizable to actual organizational or field settings ( i.e. the real world)?”

Question 19

what are the different validity threats

Answer 19

• History effects – unexpected events may disturb and cofound the cause-and-effect relationship
• Maturation effect – passage of time effects variables (gets tired) -> ex.when respondents get tired = internal validity threatened
• Testing effects – effects of exposure to pre/post tests
• Selection bias effect – from imperfect selection of participants -> types of participants selected for a lab experiment may be very different from the types of participants in natural setting -> generalizing is problematic = external validity threatened ( randomization or matching groups is highly recommended )
• Mortality effects – composition of group changes
• Statistical regression effects (regression to the mean) = extreme values tend to score closer to the mean after the treatment
• Instrumentation effects – changes in measuring tools = internal validity threatened

Question 20

explain Quasi-experimental designs

Answer 20

Pre-test and post-test experimental group design
-No control group
-Lack of randomness (self-selected groups)
-Easy to conduct

Post-tests only with experimental and control groups
-Pre-test not easy
-Not certain that two groups are equivalent
-Lack of randomness

Time series design
-Data on same variable collected at regular intervals
-Repeated measurements essential
-Good If no risk of reactivity
-Control group may be added

Question 21

what is the true-experimental designs

Answer 21

Pre-test and post-test experimental and control group design
- Both groups -» pre and post test
- Random groups

Solomon four group design
- Strong validity ( internal and external)
-useful if there are concerns that the post-test might be influenced by the pre-test -> yet difficult to set up
-2 groups treatment, 2 non
-2 groups pre-test, 2 non

Factorial design
-2 or more manipulations at the same time
-enables to test effect of two or more manipulations at the same time on the dependent variable -> enables to look for interaction

(double) blind studies
-Do not know if receiving treatment ( the participants)

Question 22

what is the application of true experimental design

Answer 22

for each type of true experimental design , think of two effects that might pose a threat to validity

Question 23

what are the advantages of true experimental design

Answer 23

• Appropriate for establishing causal linkage of one variable to another
  
  • Manipulation of IV is easy
  • Control group serves as a comparison
  • Pre-test and post-test allow for checking if manipulation occurred before outcome
  • Contamination can be controlled for effectively
  • Replication helps to find average effect of IV across people, time, situations

Question 24

what are the disadvantages of true experimental design

Answer 24

• Artificiality of lab setting
• Generalization from non-probability samples
• Number of variables researcher can include is more limited than in survey research
• Costly if repetition is required, compared against surveysIs study of present or future, not about past
• Factors should be easy to manipulate (↔ respondent’s education, social competence,…)
• Ethical limits

Question 25

what is true experimental design

Answer 25

involves randomly assigning participants to different conditions, manipulating an independent variable, and carefully controlling for potential confounding factors to establish a cause-and-effect relationship.