Week 5 - Research designs

Quasi-experimental disadvantages

Confident conclusions cannot always be drawn Alternative explanations may still exist• Group assignment based upon self-selection True random assignment is impossible/impractical Might be more systematic differences other than the independent variable Often relying on self reported information with may give Response biases

Sample screening

A large sample is surveyed and only individuals with specific characteristics are invited to participate in the actual study

Non random assignment

A non-random assignment of participants is the opposite. In this case, you want one group to consist only of people with a certain characteristic and the other group to consist only of people without that characteristic. An example of a non-random assignment of research participants is for testing a cholesterol drug on two groups of people. Group A consists of caucasian males, group B consists of latino males and group C consists of african-american males. We now can take the test results and see if any of these ethnic groups statistically respond differently to the cholesterol drug.

Quasi-experimental research

A quasi-experiment is designed a lot like a true experiment except that in the quasi-experimental design, the participants are not randomly assigned to experimental groups. Quasi-experiments are employed when the researcher is interested in independent variables that cannot be randomly assigned. Usually this happens when the independent variable in question is something that is an innate characteristic of the participants involved.

Case study

A single individual is studied in detail •Typically utilised in situations where the population of interest is unique or rare Individuals with damage to specific brain regions Individuals with rare diseases • Findings cannot be generalised very far

Within-subjects design

A within-subjects design is a type of experimental design in which all participants are exposed to every treatment or condition. The term treatment is used to describe the different levels of the independent variable. In other words, all of the subjects in the study are treated with the critical variable in question. For your independent variable, you decide to try two different types of exercise: yoga and jogging. Instead of breaking participants up into two groups, you have all participants try yoga before taking a memory test. Then, you have all participants try jogging before taking a memory test. Next, you compare the test scores to determine which type of exercise had the greatest effect on performance on the memory tests.

Double blind

Both the participant and researcher are unaware which condition they are in/delivering

Quasi-experimental advantages

Can also be conducted using survey methods Cheap Venues, equipment, and research assistants are not always needed Ethically viable

Longitudinal

Collecting data over a long period of time• A form of repeated-measures (within-subjects) design Collected from the same individuals across their life span Track specific changes: Developmental (e.g., neurobiological changes, language) Epidemiological (e.g., disease, health) Psychological changes (e.g., intelligence, personality)

Quasi-experiment example

Dr. Jones is a personality expert who studies the impact that personality traits have on intelligence. For the purposes of her current research project, she is interested in examining the IQ scores of people who score highly in each of the five 'Big Five' personality factors. Each of the five personality factors are a quasi-independent variable. Personality traits are inherent to each person, so random assignment cannot be used. Participants would initially be assigned to groups based on their personality assessment score across each of the five personality factors. Now that Dr. Jones has her participant group assignments, she can examine the impact that personality factors may have on intelligence. If a true experimental design were used, each participant would be randomly assigned to each personality group regardless of whether or not they possessed those personality traits, which would not really address the question that Dr. Jones is asking.

Practice effects

E.g., Participants improve on working memory task because they have done it before, not because they played Sudoku for a week in between. Can be problematic

External validity

External validity is the extent to which the results of a study can be generalized to other situations and to other people.

Field study

Field experiments are done in the everyday (i.e. real life) environment of the participants. The experimenter still manipulates the independent variable, but in a real-life setting (so cannot really control extraneous variables).

Counterbalancing

If half the participants complete the instruments in the reverse order then any order effects will hopefully cancel out

Cross-sectional

If you wanted to know if the percentage of women diagnosed with breast cancer increases with age, how would you go about answering this question? One way you could find the answer is to look at three groups of women who are similar but of different ages. Let's say your three age groups are 20-35 years, 36-50 years, and 51-65 years. You can then calculate the percentage of women in each group that have been diagnosed with breast cancer. This information can then be used to answer your question. This is an example of cross-sectional research. Cross-sectional research involves using different groups of people who differ in the variable of interest but share other characteristics, such as socioeconomic status, educational background, and ethnicity.

Between-subjects design

In the design of experiments, a between-group design is an experiment that has two or more groups of subjects each being tested by a different testing factor simultaneously.

Correlational research

Independent and dependent variables often measured at the same time. Only strength of relationship can be determined but causal relationship could be ambiguous. It is best suited to online survey data collections.

Internal validity

Internal validity refers to how well an experiment is done, especially whether it avoids confounding (more than one possible independent variable acting at the same time). The less chance for confounding in a study, the higher its internal validity is.

Hawthorne effect

Knowledge that one is being observed can lead participants to respond in systematically similar ways E.g., high heart rate during image exposure

Order effects

Participants may fall into response bias patterns when completing lots of questionnaires due to fatigue or boredom Participants might report higher depression symptoms ratings after completing a questionnaire about their alcohol use (priming effect)

Expectancy/reactance effects

Participants' pre-existing ideas about the study might lead them to respond in systematically similar way E.g., males might rate Brad Pitt's attractiveness lower because the study is about detecting homosexual tendencies.

Random assignment

Random assignment or random placement is an experimental technique for assigning human participants or animal subjects to different groups in an experiment (e.g., a treatment group versus a control group) using randomization, such as by a chance procedure (e.g., flipping a coin) or a random number generator.

Random selection vs random assignment

Random selection refers to how sample members (study participants) are selected from the population for inclusion in the study. Random assignment is an aspect of experimental design in which study participants are assigned to the treatment or control group using a random procedure.

Response bias

Response bias (also called survey bias) is the tendency of a person to answer questions on a survey untruthfully or misleadingly. For example, they may feel pressure to give answers that are socially acceptable.

Threats to internal and external validity

Sampling (non-random assignment) •Researcher control •Attrition (repeated measures or longitudinal studies) •Practice effects (improvements due to exposure)

Blind

The participant or the researcher do not know which condition they are in/delivering

Confounding variables/covariates

Variables that may also affect dependent variable.

Dependent variable

Variables which change in response to an independent variable

Stratified sampling

a sample which has proportions reflecting the population of interest i.e., in a population of interest there are 70% of married persons and 30% unmarried persons, thus a researcher might ensure their sample has the same proportions of married and unmarried individuals if marital status is the psychological construct of interest

Random sampling

each individual in the population has equal chance of being selected offers best opportunity to generalise to the wider population impossible to achieve in practice

Purposive sampling

individuals are deliberately chosen to represent population of interest E.g., depressed individuals can be generalised to similar populations

Convenience sampling

the most commonly used sample E.g., volunteers in the near vicinity undergraduate psychology students are the most common example offers poor generalisation to the wider population (e.g., external validity)

Independent variable

variables which cause changes in another variable