Quantitative Research Methods Flashcards
(54 cards)
The experiment: definition
aim to provide causal evidence for a relationship by showing the result of manipulating a particular factor
–> independent variable (Manipulation vs control) : Gamification vs Baseline
–> Dependent variable is measure : Actual purchase (objective) / Satisfaction (subjective)
Core elements of an experiment
- Manipulation
- Randomization
- Measurement
- Control
Direct causality
X causes Y
–> The more you study, the better your grades
Reverse causality
Y causes X
–> The faster the windmills turn, the more wind is observed
Common cause
X and Y are consequences of a common
cause, but do not cause each other
–> Sleeping with your shoes on and waking up with a headache.
Bidirectional or cyclic causality
X causes Y, and Y causes X
–> The number of lions affects the number of gazelles, but the number of gazelles also affects the number of lions.
Indirect causality
X causes Z, which causes Y
–> Unsatisfied employees are more likely to leave the company, leading to a higher turnover rate. A higher turnover rate in turn
leads to lower job satisfaction among employees.
No causality
There is no correlation between X and Y;
the correlation is a coincidence
–> Increased chocolate consumption increases the likelihood of
winning a Nobel Prize
Random assignment of participants to conditions
By randomly assigning participants to an
experimental condition, all uncontrolled
attributes (e.g., personality) of the entire
population are evenly distributed across all
conditions
–> confounding variables of an individual are eliminated by a randomized sample
Random assignment to conditions
- Control group
- Manipulation 1
- Manipulation 2
Experimental designs
- Between-Subjects
- Within Subjects
Between Subjects
Different participants are assigned to different groups, with each group experiencing only one condition of the independent variable
Within Subjects
The same participants are exposed to all conditions of the independent variable
Between Subjects: advantages
- No carryover effects (e.g., fatigue, learning).
- Participants are not influenced by previous conditions
Between-Subjects Design: disadvantages
- Requires more participants
- Greater risk of individual differences skewing results (unless groups are well-matched or randomized)
Within-Subjects Design: Advantage
- Fewer participants needed
- Reduces variability caused by individual differences
Within-Subjects Design: disadvantages
- Risk of order effects (e.g., learning, fatigue)
- Can lead to demand characteristics (participants guess purpose)
Advantages of laboratory experiments
- High internal validity
- Greater control of variables
- Replication of results
- Complex measurements possible
(brain activity, eye tracking, etc.)
Disadvantages of laboratory experiments
- Artificial environment
- Reduced generalizability
- High reactivity
- Low external validity
Advantages of field experiments
- High external validity
- High representativeness
- Low reactivity
- Natural environment
Disadvantages of field experiments
- Problems with data protection
- Lack of control over variables
- High costs
- Replication of results
Advantages of online experiments
- Global reach
- Speed of data collection
- Replication of results
- Low cost
Disadvantages of online experiments
- Distorted attributes of the participants
- Software skills
- No control over time and place
- Difficult to answer questions
Quasi-experiment
- No random assignment to the experimental
conditions - Not possible or practical to control for all
potential confounding variables
➜ Causality can only be determined to a
limited extent
