03 experimental research 2

Question 1

What are confounding vairables?

Answer 1

Confounding variables are variables that are not the primary focus of the study but may interfere with the analysis, leading to incorrect conclusions about the relationship between the variables of interest

Question 2

What is an experiment?

Answer 2

is taken to mean a scientific investigation in which an investigator manipulats or controls one or more independent variables and observes the dependent variables for variation concomitant to the manipulation of the independent variables

Question 3

What is the main idea behind experiments?

Answer 3

we want to manipulate our wordl in order to learn its secrets

Question 4

What is the distinction of experiments in regards to other methods?

Answer 4

In experiments we have a active manipulation of the independent variables that is the focal distinction to other methods

Question 5

What kind of techniques exist to control for confounding variables?

Answer 5

Control for confounding variables:
- Controlled conditions
- Experimental laboratories
- Control groups
- Random assignment to experimental conditions

Question 6

Two forms of experimental studies to detect relations between two constructs or variables?

Answer 6

1. Non-experimental: A and B are observed (external validity) “observe nature in raw” -> no intervention
2. Experiments: A is manipulated and effect on B observed (internal validity)

Question 7

The two types of confounds?

Answer 7

Respondents-related:
- Age
- Income
- Education
- Motivation
- Memory
- Intelligence

Situation-related:
- Time of study
- Researcher
- Location of study
- Nationality
- Light

Question 8

What is randomizatiion?

Answer 8

Controlling for respoondents-related confounds via randomization

randomization: controls for the effects of observable/unobserverable characteristics and makes group equivalent (on average) on all variables

–>it helps to assess whether differecnes after treatment are larger than would be expected due to chance

Question 9

What does it mean if we have random asisngment in an experiement?

Answer 9

Random assignment in two groups:
1. 50 get daily vitamin C tablets
2. 50 get daily placebo tablets

–>because of this random assignment, we can make the reasonable assumption that any differences you observe is only due to the manipulation

Question 10

Can we use experiments to inferre causal relationships?
(4 coniditions)

Answer 10

1. Dinstinct entities: depends on the conceptual considerations, no specific experimental mechanism
2. Association:Yes–>DV are observed for different levels of IV
3. Temporal precedence: Yes –>DV are observed temporarily after the manipulation of IV
4. Eliminating rival causal explaination: Randomization creates structual equality for all variables that are not manipulated, also use confound checks and process measure possible

Question 11

Process of conducting experimental studies?

Answer 11

1. selection of research variable
2. Choice of experimental design
3. Creating the experimental stimuli
4. Data collection
5. Data analysis
6. Presentation of findings

Question 12

What ar the 3 aspects to consider by designing an experiment

Answer 12

environment setting:
- Field experiment (realistic) vs. Laboratoy experiment (Physical re-enactment/ scenario description)

randomization:
- Random experiment (Complete) vs. Quasi-experiment

Point of reference:
- Within subjects design (same participants) vs. between subjects design (different participants –>better)
–>+ N-factorial design

Question 13

What are possible reasons for drawing false causal inferences in experiments:
External validity threats

Answer 13

Sample selection bias: occurs when the sample selected for the study is not representative of the larger population
–>Impact: Findings may not be generalizable to the broader population

Low group sizes: –>related to statistical power
–>Impact: results may lack generalizability, more vulnerable to random variation
—> larger sample size, provides more reliable and representative data

Question 14

What are the 8 threats for interanl validity?

Answer 14

1. Group threats (selection bias)
2. History
3. Testing effects
4. Regression to the mean
5. Maturation
6. Mortality
7. Instrument change
8. Reactivity and experimenter effects

Question 15

What is: Group threats, history and testing effects? (Internal validity threats)

Answer 15

Group threats (selection bias): occur when you do not randomize and let peiple decide on the group

History: external events occuring during the study that could affect the outcomes
–>Impact: discounts at the same time, might have an impact on the results)

Testing effects: Important for repeated measurements: Changes in participants behavior due to the act of being tested or assed rather than the experimental treatment itself

Question 16

What is: regression to the mean, maturation, or mortality?

Answer 16

Regression to the mean: the tendency of extremce scores on a variable to move toward the average on subsequent measurements

Maturation: Changes in participants over time that are not a direct result of the experimental treatment but simply natural maturation, for i.e. change behavior due to conducitng multilpe experiments

Mortality: loss of participants from a study over time

Question 17

What is instrument change, reactivity and experimenter effects?

Answer 17

Instrument change: Alterations in the measurement instruments or tools used during the study

Reactivity and experimenter effects: unitentional biases or cues from the experimenter that influence participants behavior

Question 18

What is a field experiment?

Answer 18

Field experiment:
- Takes place in a realistic setting
- Subjects are mostly unaware of their participation in an experiment

Question 19

What is an Laboratory experiment?

Answer 19

Laboratory experiment:
- Takes place in environment created especially for the experiment
- Subjects are aware of their participation in the experiment (often actual research question is hidden)

Question 20

What are the advantages of a field experiment?

Answer 20

Advantages field exppeirment:
- high degree of authenticity increases the possibility of making generalizations about the experimental findings (high external validity)
- No awareness of participation

Question 21

What are the disadvantages of an field experiment?

Answer 21

Disadvantages:
- Very expensive
- ethically problematic
- Barely replicable (decreases generalizability)
- Little control over potential confounding variables (low internal validity)
- Manipulation of independent variables often very difficult
- Access to organizations often difficult

Question 22

What are the advantages of laboratory experiments?

Answer 22

• Time and costs
• High control over potential confounding variables (high internal validity)
• Easy to replicate
• High level of control over manipulation of the independent variables

Question 23

What are the disadvantages of laboratory experiment?

Answer 23

Disadvantages laboratory:
- Little authenticity (low external validity)
- Subjects are aware of taking part in an experiment

Question 24

What are the two forms of laboratory experiemnts? (Also describe)

Answer 24

Physical re-enactment:
- exerpimental conditions are physicall re-enacted in the experimental laboratory
- COmbiens survey methods and observational methods

Scenario description:
- subjects are asked to image themselves in a certain fictional situation and then to describe their own behavior
- specific form of survey methodology

Question 25

What are the advantages of a phsical re-enactment?

Answer 25

**Advantages phycisal re-enactment**: - **Higher degree of authenticity** **increases** the possibility of making **generalizations** about experimental findings (**higher external validity**) - **Observation of physical** reactions (especially behavior) possible

Question 26

What are the disadvantages of physical re-enactment?

Answer 26

**Disadvantages**: - Relatively **costly** - Many organizational **situations cannot be recreated** in a lab (e.g., decisions about high-priced purchases) - **Sometimes ethically ambivalent** (e.g., when researching negative experiences)

Question 27

What the advantages of scenario descriptions?

Answer 27

**Advantages scenario descriptions**: - **Less costly** (can be conducted online) - **Possibility to examine negative experiences**

Question 28

What are the disadvantages of scenario descriptions?

Answer 28

**Disadvantages scenario descriptions**: - **Little authenticity** (**low external validity**) - **Scenarios cannot be very detailed** – high degree of subjective assessments by the participants - **Behavior cannot be readily observed** → risk of **Common Method Effects** - Usually **restricted to visual stimuli**

Question 29

What is the difference between quasi-experiments and randomized experiments?

Answer 29

**Quasi-experiments**: experiments that **lack random assignment of units to conditions** but that otherwise have similar purposes and structural attributes to randomized experiments **Randomized experiments**: randomly allocating the subjects across the **experimental conditions** (**randomization**

Question 30

What is the difference between random assignment and random sampling?

Answer 30

DIfferent **objectives**: **random sampling**: structural equivalence of sample and population **Random assignment**: structural equivalence between the **different experimental conditions** (groups)

Question 31

What is the **goal of randomization**?

Answer 31

**Goal randomization**: - Alternative causes **are not confounded with the treatment** - **Reduces the liklihood** of **threats to validity by randomly distributing** them across experimental conditions - All experimental conditions have the **same expected** value of all variables, measured or not - Allows the researchers to know and model the selection process correctly - Allows compuation of **valid estimate of error variance** that is **uncorrelated** with the experimental treatment

Question 32

What is a facortial design?

Answer 32

**Factorial design**: - dominant in business research - **several IV** (factors) each with **at least two attribute levels** - each combination of the **factor levels** corresponds to **one experimental condition (cell)** Notation in case of two IV with two levels each: R: randomization, X: treatment, O:observation Number of **experimental conditions** (Cells): - product of factor attribute levels (crossed design) - Can lead to **large designs** - Possibility of **Nested designs**: each attribute level of one factor is only combined with a specific level of the other factor

Question 33

What are the advantages of **factorial design**

Answer 33

**Advantages of factorial design**: - smaller sample sizes (each particippant exposed to several IV) - allows for **testing interactions** between the **IV**

Question 34

What are the important factors by **designing the experimental stimuli**?

Answer 34

1. **realism**: how closely the experimental stimuli mirros the **real-world** 2. **Unobtrusiveness (**unauffdringlichkeit**) of the issue being research**: refers to the degree to which the experimental stimuli avoid interferring with or alteirng natural behavior 3. **Simplicitiy** involces the clarity, straightforwardness, and lack of uncessary complexit

Question 35

What are the 3 stage of the **experimental manipulation** process? (Design of experimental stimuli)

Answer 35

1.**Treatment delivery**: researcher privde the specified treatment (stimulli, intervention, condition) 2.**treatment receipt**: Researcher asses whether participants **exposed** to and excounter the treatment 3.**Treatment adherence**: assess whether pariciants **comply with instructions**, requirements or protocols

Question 36

How to check the **effeectiveness of manipulation** in the design of experimental stimuli

Answer 36

**Manipulation checks**: examine wheethe the manipulation sucessfully influenced the varibale of interest **Realism checks**: participants are asked about perceived authenticity or believability of experimental setup **Confound checks**: (assess whether any **unintended** influences or confounding varibales may affect the experimental manipulation

Question 37

Sample size and sample structure: What can be down against the problem: **participants** drop out? (Risk and solutions)

Answer 37

**Problem: participants drop out** - especially serious in case of overlap with research question - **Risk**: random experiments convert into quasi-experimetns **Solutions/options**: - minimize the amount of tme between agreement to participate, manipulation, and measurement - Get consent of participants to participate in all possible experimental conditions - Analyze

Question 38

What are the ethical principles of experimental research?

Answer 38

**Informed Consent:** – Inform participants about experimental content – Participants always have the opportunity to end the experiment – In case of observations/field studies: Protection of personal rights/privacy **Deception:** – Truthful instructions – No misguidance of participants – No active support of failure **Debriefing:** – After conducting the experiment, inform participants about the goal of the study **Confidentiality:** – Protection of personal rights/privacy – Analysis of results on an aggregate level **Protection from physical and psychological harm:** – Avoidance of stress, fear, shame etc. – Avoidance of physical damage