L8 - Experimental design

Question 1

Objective of an experiment

Answer 1

To determine a causal link between two or more variables.
» stronger evidence than exploratory or descriptive designs because of the control made possible by causal research designs.

Question 2

2 pathways of the experiment design

Answer 2

The “Between” or “Within” subjects approach.

Question 3

‘Between’ subjects experiment design

Answer 3

Multiple samples - only one single condition
> Pros: Require less time.
> Cons: the need to ensure the comparability of results regarding the differences in demographic.

Question 4

‘Within’ subjects experiment design

Answer 4

One sample - multiple single conditions
> Pros: Not require above examination of demographics consistency; Higher detection of effects as no possibility of extraneous subject variable; Ongoing record of subject’s behaviour over time
> Cons: Potential bias; Cannot get many subjects.

Question 5

Three approaches for analyzing sample data that are based on the number of variables?

Answer 5

Univariate, Bivariate and Multivariate

Question 6

Univariate data analysis

Answer 6

• The type of data in which analysis are made only based on one variable.
• Variables are examined individually and in isolation with no interactions examination.

Question 7

Bivariate data analysis

Answer 7

• Analysis based on two variables per observation simultaneously.
• Cross tabulation technique enables exploration of relationships between 2 variables.

Question 8

Multivariate data analysis

Answer 8

• Analysis based on three or more variables per observation.
• It is used for explanatory purposes.
• Advanced data analysis techniques like ANOVA, Regression modeling, and more.

Question 9

ANOVA (analysis of variance)

Answer 9

Analyses and compares the variances of datasets to conclude whether the distribution of variance between different data sets in similar, demonstrating a relationship between data or not.

Question 10

1-way ANOVA

Answer 10

1 dependent variable and only 1 factor regardless of the amount of conditions.
Ex: 1 factor = colour; 3 conditions = blue, yellow and red.

Question 11

2-way ANOVA

Answer 11

1 dependent variable and 2 factors in it.

Ex: Colour = blue, yellow, red; Length = short, medium, long)

Question 12

Social desirability bias

Answer 12

The participants’ tendency to describe themselves in favourable terms by adhering to socio-cultural sanctioned norms.

Question 13

Motive of socially desirable responding

Answer 13

• To obtain approval by responding in a culturally appropriate and acceptable manner.
• To portray themselves in a more positive light when answering sensitive topics.
• To impress the questioner, an unwillingness to admit certain behaviour or attitudes, or an attempt to influence the outcome of a study.

Question 14

The drivers of Social desirability bias (Fisher, 1998)

Answer 14

• This includes: Impression management, Self-deception, Level of involvement, Extrinsic benefit, Intrinsic benefit, Social norm influence.
• It depends on the particular sets of circumstances.

Question 15

Indirect questioning (Fisher, 1998)

Answer 15

It is a projective technique that asks respondents to answer structured questions from the perspective of another person or group.
It assumes that respondents project their unconscious biases into ambiguous response situations and reveal their true feelings; or giving information about situations based on fact rather than opinion and so they respond behind “a facade of impersonality”.

Question 16

Limitation of indirect questioning in reducing SDB (Fisher, 1998)

Answer 16

• The degree of similarity between the respondent and the typical other decreases, the level of irrelevant information in their answers increases. Ex: non-drinking person are less likely to project themselves as heavy consumer of alcohol.
• Predictions about others do not accurately reflect what respondents would have reported had they been asked directly.

Question 17

Implications of Indirect questioning in reducing SDB (Fisher, 1998)

Answer 17

• SDB would have a significant effect on directly worded items but not indirectly worded items.
• Indirect questioning provides a better estimate of the true scores of socially-sensitive variables.

Question 18

Lying behaviour of consumers

Answer 18

When there is a gap between ‘the perceived cost of providing personal info and the benefit perceived to be received in return’, greater likelihood for the individual to lie.

Question 19

How to tackle with SDB and lying behaviour

Answer 19

Randomized response methodologies - ensuring full anonymity for all respondents.

Question 20

Control variables

Answer 20

• Variables that the researcher control. Ex: control prices when test the influence of package design on sales.
• Don’t affect the functional relationship between the independent and dependent variables in the experiment.

Question 21

Extraneous variables

Answer 21

• Uncontrollable variables that should average out over a series of experiments. Ex: mood, moistness on cake.
• Can have a confounding impact on the dependent variable measures that could weaken or invalidate the results of an experiment.

Question 22

Internal validity

Answer 22

The extent to which the research design accurately identifies causal relationships. It refers to whether the manipulation of the independent variables or treatments actually caused the effects on the dependent variables.

Question 23

External validity

Answer 23

• The extent to which the results of the experiment can be generalized to the target population.
• Random sampling is necessary for EV, but it is not sufficient to confirm the study can be generalized.

Question 24

Field experiment (slide)

Answer 24

Non-controlled settings where experiments are conducted based on observation of the assigned group over a period of time.

Question 25

Problems of Field experiment (slide)

Answer 25

• High realism: hard to control extraneous and independent variables.
• Longer timeframe time to complete.
• High cost.
• Competitive reactions: risk of being copied.

Question 26

Laboratory experiment (slide)

Answer 26

• Controlled environments where the researcher is present and monitors data collection to conduct casual research designs.
• However, it lacks of external validity.

Question 27

Advantages of Laboratory experiment (slide)

Answer 27

• High control; Short time, Easy to implement, Low cost.

- High internal validity.

Question 28

Limitations of Experiments (slide)

Answer 28

• Impact of confounding factors.
• Internal and External validity.
• Time consuming when it measures long-term effect.
• Expensive due to requirements of experimental group, control group, and multiple measurements.
• Difficult to administer: may be impossible to control the effects of extraneous variables.

Question 29

Reactive error (slide)

Answer 29

The change in normal behaviour due to the manipulation condition. It is because respondents may understand what the researchers are trying to do.

Question 30

Post hoc test

Answer 30

• It is necessary to follow up to any ANOVA. Since ANOVA allow identification of significant results, meaning there is a relation leading to causality, but does not specify what kind of relationship this is.
• Ex: Tukey, LSD, Scheffe