Research Methodology - Exam 3

Quantitative Variables

numerical values that can also be counted or measured.

Categorical Variables

describe qualities or characteristics that represent groups or categories.

Mehl et al.'s (2010) Study on Well-Being and Deep-Talk

Purpose: The study investigated whether happy people have conversations differently from less happy people. It examined the relationship between well-being and the amount of small talk or deep conversations people have.

- Each participant wore an EAR, a device that recorded 30-second snippets of sound every 12.5 minutes over four days.

- Participants also did self-report measures about life satisfaction and happiness.

Conclusion: Those who had deeper, meaningful conversations = higher well-being scores.

Siddarth et al. (2018) Study on Sitting and MTL Thickness

Medial Temporal Lobe

Purpose: The study investigated whether sitting time is associated with medial temporal lobe thickness in adults. The MTL is critical for memory (HIPPOCAMPUS!!).

- Asked the adults about their average sitting time and physical activity. Then, they used MRI scans of the brain to measure MTL thickness.

Meeting Location and Marital Satisfaction

Purpose: To investigate where couples meet and their marriage satisfaction. Online or in-person meeting.

- 19,000 respondents who completed an online questionnaire about their relationship and marital satisfaction. It was measured on a 7-point scale.

Construct Validity

How well was each variable measured?

- Mehl Study: How well were each of the variables measured?

- Does the measure have good reliability?

- Is it measuring what it's intended to measure?

- What is the evidence for its face validity, for its concurrent validity, and for its

discriminant and convergent validity?

Statistical Validity (Effect Size)

Describes the strength of an association between two or more variables.

- The r value being closer to 1 equals a stronger correlation.

Confidential Interval (CI): Large sample vs. Small sample

The probability that a parameter will fall between a set of values, expressing the certainty in a percentage.

- A larger sample size makes it narrower and produces more precise confidence intervals.

- A smaller sample size makes it less stable, wider, and less precise confidence intervals.

Replication

Has the study been conducted again? To test if the result is consistent.

Outliers

One or more extreme scores that lie far away from the rest of the data.

Restriction of range

When there is not a full range of scores on one of the variables in an association in a correlational study, it can make the correlation appear smaller than it truly is.

- Ex) SAT scores and college GPA.

Curvilinear Association

One in which the correlation coefficient is zero (or close to), and the relationship between two variables isn't a straight line.

- Ex) Age and the use of health care services - as people get older, their usage of health care decreases, and then eventually, when they reach around 60, it increases.

Covariance

There must be an association between the causal variable A and the effect variable B.

Temporal Precedence

The causal variable A must come before the effect variable B.

Third Variable Problem

Is there a third variable C that is associated with variables A and B independently?

External Validity: How important?

Random sample is important because it should NOT cause us to automatically reject the association. We can accept the results as is. Save the question of generalizability for another study that tests the same variables in a different population.

Moderating Variable

When the relationship between two variables changes depending on the level of another variable.

-Ex) Correlation between professional sports game attendance and the success of the team; you might expect to see a positive correlation, the more a team wins, the more people will attend. HOWEVER, this association is MODERATED by the franchise city's level of mobility (how often people move). If the city has low mobility, people grow more ties and go to games, and vice versa.

Longitudinal Design

Helps address temporal precedence. Can provide evidence for temporal precedence by measuring the same variables in the same people at several different times.

- Ex) Framingham Heart Study: focused on what causes heart disease, stroke, etc. How can we cut the risks? Generational health history, lifelong. ONE OF THE BEST LONGITUDINAL STUDIES!!

-Ex) Murdock Study: Serial position effect; found that people are better at recalling items from the beginning (primacy) and at the end of a list (recency).

Cross-Sectional

Tests of whether two variables measured at the same point in time are correlated. (X and Y)

Autocorrelation

The correlation of each variable with itself across time.

Cross-lag correlation

Correlations of the degree to which an earlier measure of one variable is associated with a later measure of other variables. Examines how people change over time.

Why not just do an experiment?

In many cases, participants cannot be randomly assigned to a variable. People cannot be ethically or practically manipulated. It may be impossible or unethical to assign participants.

Quasi-Experiments

In a quasi-experiment, researchers do not have full experimental control. The researchers might not be able to randomly assign participants to one level or the other; they are assigned by teachers, political regulations, acts of nature, or even by their own choice.

Four Examples

1) Nudging people towards organ donation.

2) The psychological effects of cosmetic surgery.

3) Popular shows and suicide.

4) Investigating the effect of legislation on opioid abuse.

What is a quasi-independent variable?

A variable that resembles an independent variable, but the researcher does not have true control over it.

What is an example of a situation where a quasi-independent variable is used?

When the researcher cannot randomly assign participants to its levels or cannot control its timing.

What is a participant variable?

A categorical variable, such as age, gender, or ethnicity, whose levels are measured instead of manipulated.

Nonequivalent control group posttest-only design

A quasi-experiment that has at least one treatment group and one comparison group, in which participants have not been randomly assigned to the two groups, and in which one pretest and one posttest are administered. The participants were only tested once, after exposure to one level of the independent variable or the other.

- Ex) Nudging people toward organ donation.

Nonequivalent control group pretest/posttest-only design

A quasi-experiment that has at least one treatment group and one comparison group, in which participants have not been randomly assigned to the two groups, and in which one pretest and one posttest are administered. Since participants were not randomly assigned to groups, and were tested both before and after some intervention.

- Ex) The psychological effects of cosmetic surgery.

Interrupted time-series design

A quasi-experiment in which participants are measured repeatedly on a dependent variable before, during, and after the "interruption" caused by some event.

- Ex) Popular shows and suicide.

Nonequivalent control group interrupted time-series design

A quasi-experiment with two or more groups in which participants have not been randomly assigned to groups; participants are measured repeatedly on a dependent variable, before, during, and after the "interruption" caused by some event, and the presence or timing of the interrupting event differs among the groups.

- Ex) Investigating the effect of legislation on opioid abuse.

What is the wait-list design?

An experimental design for studying a therapeutic treatment, in which researchers randomly assign some participants to receive the therapy under investigation immediately, and others to receive it after a time delay. The wait-list design ensures that the same kinds of people are in each group.

Maturation threat

Maturation threats occur when, in an experimental or quasi-experimental design with a pretest and posttest, an observed change could have emerged more or less spontaneously over time. This can be fixed with a control group.

- Ex) cosmetic surgery

History Threat

A history threat occurs when an external, historical event happens for everyone in a study at the same time as the treatment. With a history of threats, it is unclear whether the outcome is caused by the treatment or by the external event or factor.

- Ex) Did a celebrity kill themselves that month?

Attrition Threat

Attrition occurs when people drop out of a study over time. Attrition becomes an internal validity threat when systematic patterns of people dropping out of a study occur.

- Ex) In cosmetic surgery, the people who still had low self-esteem after the surgery may have dropped out of the study.

Testing and instrumentation threat

A testing threat is an order effect in which participants tend to change as a result of having been tested before. Instrumentation, too, occurs when participants are tested or observed twice.

H.M. (Henry Molaison) case study

A surgeon excised the front half of his hippocampus on both the right and left sides (bilaterally), along with other brain areas, including portions of his hippocampal gyrus, amygdala, and medial temporal lobes. Studies showed that Henry's short-term (working) memory was fairly intact. He could remember a set of numbers (such as "5, 8, 4") for a short period of time, reciting a span he'd heard up to 30 seconds earlier. And if he continually rehearsed a number, he could accurately recite it even an hour later. But if he was distracted for a minute (talking to a nurse, for instance), it was lost. When asked to recall it, he'd say, "What number?"

Ebbinghaus

He was the first to document several fundamental memory phenomena, one of which is called the forgetting curve, shown in Figure 13.17. It depicts how memory for a newly learned list of nonsense syllables declines most dramatically over the first hour but then declines more slowly after that. Although he studied only himself—a single-N design—Ebbinghaus created testing situations with experimental precision and, eventually, replicability.

Piaget and Child Development

He found, for example, that, as children get older, they learn that when a short glass of liquid is poured into a smaller glass, the amount of liquid doesn't change. Younger kids, in contrast, tend to believe that a taller glass is because the level appears higher. Although Piaget observed only a few children, he designed systematic questions, made careful observations, and replicated his extensive interviews with each of them. HE USED HIS OWN THREE KIDS!!

Stable-baseline design

A researcher observes behavior for an extended period before beginning treatment or other intervention, and continues observing behavior after the intervention.

Multiple-baseline design

Researchers stagger their introduction of an intervention across a variety of contexts, times, or situations.

Reversal design

A researcher observes a problem behavior both before and during treatment, and then discontinues the treatment for a while to see if the problem behavior returns.

a Single-N design

Researchers gather information from only one animal or one person.

Internal Validity

The internal validity of these single-N studies is high.

External Validity

It may seem easy to criticize external validity. Remember, though, that researchers can take steps to maximize the external validity of their findings.

Construct Validity

Construct validity is fairly straightforward when researchers are recording whether a person with a brain lesion reports seeing this picture or that one, and for objective measures such as memory for a string of numbers.

Statistical Validity

Researchers typically use traditional statistical methods. However, they still draw statistical conclusions from data, and they should treat data appropriately.