Research Methodology Exam 3

Bivariate Correlations

Associations that involve exactly 2 variables

When examining an association in which one variable is categorical and one is quantitative, which of the following is most likely to be used?

a bar graph

True or False- When examining an association claim using a bar graph, an association is indicated by a difference in the height between the bars.

True

Which of the following is true of 95% CIs?

It deals with precision estimates about the population

Which of the following is true about outliers?

They have the biggest effect when dealing with small sample sizes

If an association study did not select people for the study by using random sampling, which of the following statements is true?

The findings should be replicated in another population.

Which of the following questions is necessary to ask when interrogating statistical validity?

Is there a restriction of range?

A study finds a correlation coefficient of r =.52. This number gives you information about which of the following?

strength and direction of the relationship

Which of the following indicates that a study used a bivariate correlational design?

the presence of two measured variables

A study finds a correlation coefficient of r =.52 and reports 95% CI [.37, .67]. The 95% CI is a

confidence interval

When bivariate association claims do not meet the criteria of temporal precedence and internal validity, this means that cannot be .

causal inferences; made

True or false- Replicating a study gives the researcher additional estimates of the association

True

Medial Temporal Lobe Convergence zone

Includes hippocampus, important for declarative memories

Construct validity

How well was each variable measured?

Does the measure have good reliability?

Is it measuring what it is intended to measure?

Statistical Validity

How well does the data support the conclusion?

How strong is the relationship?

How precise is the estimate?

Could outliers affect association?

Effect size

strength of an association (large effect sizes are more important

Outlier

one or more extreme scores that lie far away from the rest of the scores

Restriction of range

When there is no full range of scores on one study, it can make the correlation seem smaller than it really is

(R=Pearson Correlation)

Internal validity

Can we make causal inferences from the association?

External Validity

extent to which we can generalize findings to real-world settings

Curvilinear association

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

3 causal criteria

Covariance, temporal precedence, and internal validity

Covariance

There must be an association between the cause variable and the effect variable

Temporal precedence

The causal variable must come before the effect variable

Internal validity (causal criteria)

Is there a third variable that is associated with variables A & B independently? Yes= cannot infer causation

Moderator

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

Multivariate designs

Involve more than 2 measured variables (get closer to causality than bivariate designs)

Longitudinal study

addresses temporal precedence, provides evidence by measuring same variable in same participants over different times.

Murdock Study

a long-term initiative for disease reclassification through advanced biomarker discovery and integration with electronic health records

Framingham Study

A cohort study on cardiovascular disease, which was begun in 1948 in Massachusetts and consisted of a little over five thousand subjects.

Cross sectional correlations

two variables measured at the same point in time to see if they are related (Cannot establish temporal precedence to see if one happened before another)

Autocorrelations

the correlation of each variable with itself across time

Cross-lag correlation

Correlations of the degree to which an earlier measure of variable 1 is associated with the later measure of variable 2

Quasi experiment

An experiment in which investigators make use of control and experimental groups that already exist in the world at large. Also called a mixed design.

Four examples of Quasi exp from textbook

Nudging people towards organ donation

Psychological effects of cosmetic surgery

Popular show (13 reasons why) and suicide

Investigating effect of legislation on opioid abuse

Nudging people towards organ donation

Countries that had an opt-out had higher organ donor rates than those who had opt-in

Psychological effects of cosmetic surgery

Compared self-esteem of people who underwent cosmetic surgery and people who only considered it

Popular show (13 reasons why) and suicide

Looked at spikes in popularity in show with spikes of suicide rates of teenagers

Investigating effect of legislation on opioid abuse

Florida introduced opioid legislation laws that ultimately brought down opioid use in Floridians in comparison with NC who did not pass legislation and had relatively same amount of users.

Internal validity in quasi experiments

Selection effects

Design confounds

Maturation threat

History threat

Regression to the mean

Attrition Threat

Testing and instrumentation threat

Selection effects

unaccounted for differences between groups

Design confounds

extraneous variables that may influence study's results

Maturation threat

an observed change could have emerged more or less spontaneously over time

History threat

an external, historical event happens for everyone in a study at the same time as the treatment

Regression to the mean

occurs when an extreme finding is caused by a combination of random factors that are unlikely to happen in same combo again

Attrition threat

occurs when people drop out of a study over time (longitudinal studies)

Testing and Instrumentation threat

participants are not measured in exactly the same way. Participants responses change as a result of having been tested before (comparison group)

Balancing priorities in a quasi experiment

Real world opportunities, ethics, and construct validity and statistical validity

Real world opportunities

quasi-experiments present real-world opportunities for studying interesting phenomena and important events

Ethics

Many questions of interest to researchers would be unethical to study in a true experiment

Construct and Statistical validity (quasi exp)

Usually show excellent construct validity (How large group was = statistical significance of findings)

Which of the following studies is an example of a longitudinal design?

Dr. Benson's study in which she measured people's spatial manipulation ability in August and measured their ability again in May after they had taken two semesters of art classes.

Which of the following is a necessary component of a longitudinal design?

measuring the same variables at two distant points in time

Which popular media headline does NOT suggest that a multiple regression has been used?

"Dog ownership decreases stress."

All of the following are true of betas and correlation coefficients EXCEPT

betas describe the relationship between two variables exactly as correlations coefficients do.

The pattern and parsimony approach to causation is a good example of which cycle in research?

theory-data cycle

If an experiment cannot be done for practical or ethical reasons related to manipulating the variable of interest, which of the following events should happen?

A longitudinal correlational design could be done instead.

Tests of whether two variables measured at the same point in time are correlated.

Cross-sectional correlations

Correlations of the degree to which an earlier measure of one variable is associated with a later measure of the other variable

Cross lag correlations

The correlation of each variable with itself across time

Autocorrelations

The degree to which a good scientific theory provides the simplest explanation of some phenomenon.

Parsimony

Jennifer is a developmental psychologist who studies changes over time in intelligence by studying the same people at ages 16, 22, and 30. What type of research design is she using?

longitudinal

True or false- Multivariate designs will get us closer to causality than did bivariate designs.

True

Which of the following is true of a quasi-experiment?

In a quasi-experiment, the researcher assigns participants to conditions based on the particpant's preexisting level of the independent variable

Which of the following topics would be especially well suited to a quasi-experimental design?

Do people diagnosed with a mental illness have poorer social abilities?

In quasi-experimental designs, the researcher does not have experimental control over the independent variable and does not randomly assign participants to conditions. This results in which of the following?

a weaker causal claim than a true experiment

Which of the following is a quasi-experimental design in which participants are not randomly assigned to groups and are tested only after exposure to the quasi-independent variable?

nonequivalent control group post-test only design

Which of the following is a quasi-experimental design in which participants are not randomly assigned to groups and are tested both before and after exposure to the quasi-independent variable?

nonequivalent control group pretest/posttest design

Which of the following is true of small-N experiments?

Each person in a small-N design is treated as a separate experiment.

A small-N design that involves providing treatment and then removing treatment is known as a(n)

reversal design

Researchers may be interested in how a variable changes over the course of a major event that is scheduled outside of experimental control. This is called

interrupted time-series design

Nonequivalent control group posttest-only design

participants in one group are exposed to a treatment, a nonequivalent group is not exposed to the treatment, and then the two groups are compared with no pre-treatment measurement

nonequivalent control group pretest/posttest-only design

a treatment group that is given a pretest, receives a treatment, and then is given a posttest. But at the same time there is a nonequivalent control group that is given a pretest, does not receive the treatment, and then is given a posttest.

Interrupted time-series design

a quasi-experimental research method where data is collected over time, and then an intervention is introduced, "interrupting" the time series

Nonequivalent control group interrupted time-series design

taking a set of measurements at intervals over a period of time both before and after an intervention of interest in two or more nonequivalent groups.

Wait list design

a method where participants are randomly assigned to either a treatment group (receiving the intervention) or a control group (placed on a wait-list).

stable-baseline design

a researcher observes behavior for an extended baseline period before beginning a treatment or other intervention; if behavior during the baseline is stable, the researcher is more certain of the treatment's effectiveness

multiple baseline designs

researchers stagger their interventions across situations, times, or contexts

Reversal designs

researchers usually observe a baseline of behavior before treatment next observe the behavior during treatment, and then remove the treatment to see whether the behavior reverts back (reversal period)

Human memory small N

Henry Molaison and his seizures. Surgeon rid half of the hippocampus on both right and left sides. HM had mild retrograde amnesia and severe anterograde amnesia. Hippocampus essential in forming new LTM.

Single N design

one individual is tested repeatedly on each experimental condition

Ebbinghaus

created the forgetting curve (learning curve) and serial position effect in memory

Piaget

theorist that developed a series of stages in which an individual passes during cognitive development.

internal validity in small N

can be very high if the study is very carefully designed

external validity in small N

can be problematic depending on the goals of the study