Research designs (InThinking)

Independent measures designs have the problem of participant variability.  What does this mean?

• Participants drop out of the study because of boredom.

• Differences in the traits and past experiences of the participants in each condition may affect the results of the study. 

• Participants have different levels of skill, so some understand the directions better than others.

• Participants are not the same as the general population and the results cannot be generalized.

Which of the following is not true of a repeated measures design?

• Fewer participants are needed than for an independent samples design. 

• Participant variables are controlled.

• Participants are less likely to show demand characteristics than in an independent samples design.

• Participants are more likely to show order effects - such as improvement of a skill through practice - than in an independent samples design.

A researcher wants to find out if talking on a hands-free mobile phone affects one's driving ability.  Participants were asked to take a driving test in a driving simulator that monitored their mistakes.  In one condition, the participants were asked simply to take the driving test.  In the second condition, the participants were asked to dial up a friend and to have a conversation while taking the driving test.  Half the group did the first condition and half the group did the second condition. Then both groups took a one-hour break before taking the test again - but in the other condition.  What is the best description of the design of this study?

• A longitudinal case study

• A counterbalanced independent samples design

• A counterbalanced repeated measures design 

• A counterbalanced matched pairs design

Why would a researcher use a matched-pairs design?

• So that each participant experiences both conditions of an experiment.

• To control for participant variability within a sample.

• To make sure that the order in which the experiment is done does not affect the results.

• To make sure that there are equal numbers of participants in each group.

Which of the following is not an order effect?

• Fatigue

• Social desirability effect:  participants give responses that are "socially acceptable."  

• Boredom with the study

• Practice effect - participants get better at the skill being studied, just because they keep doing it during the experiment.

What is the relationship between a sample and a population?

• A sample is a group that is representative of the larger pool of people from which it is drawn.

• A sample is the participants that are taken from a group to which you want to generalize the results of the study.

• Samples always come from the researcher's culture; populations are global.

• Samples are all the people possible for a study; the population is the participants that actually show up.

Answer explanation

Unfortunately, not all samples are representative of the population.  This may be because there was sampling bias.  This may be due to the type of people that volunteer for a certain type of study  - or when using an opportunity sample, the variables in that group that may not be representative of a larger population. Therefore, you need to ensure your participants represent the population they are from.

If a researcher does a study at your school and asks for volunteers, but only psychology students sign up to take part, what is the problem with this study?

• Ethnocentrism

• The sample is too small.

• It cannot be replicated.

• Sampling bias 

Answer explanation

When psychologists use a non-random sampling technique, this may result in sampling bias -  causing some members of the population to be less likely to be included than others.  This means that participant variables in the sample may not be representative and may influence the outcome of the study.

What is the problem with using WEIRD samples?

• They are not culturally and socioeconomically representative of the global population.

• It is unethical to use "weird" people in research.

• The samples cannot be replicated.

• They exclude outliers from the sample.

Answer explanation

WEIRD stands for: Western Educated Industrialized Rich Democratic.  These characteristics are not seen as representative of the global population, so results would be limited in their generalizability.

Match the following types of sampling

• sample matches population make-up

• Random

• volunteers

• Opportunity

• names from a hat

• Snowball

• whoever is available

• Self-selected

• ask your participants to ask friends

• Stratified

According to Carol Dweck's theory, students who have a fixed mindset ...

• ...listen to feedback with the goal of improving their skills.

• ...are of limited intelligence.

• ...are less likely to take risks because they fear failure.

• ...complete their cognitive development by the age of 12.

Answer explanation

Dweck argued that those who have a fixed mindset believe that their skills are a "talent."  Therefore, they do not take risks because failure would hurt their identity.