CCE: Statistics & Research Design

What is an independent variable (IV)?

The variable that is believed to have an effect on the dependent variable and is varied or manipulated by the researcher in an experimental study.

What does a p value represent?

is an index of item difficulty. the closer it is to 1.0, the easier the item since this means that a large proportion of examinees answered the item correctly. A low p value indicates a difficult item.

What does it mean when factors in a factor matrix are "oblique”?

=correlated so the correlation coefficient for any two factors is greater than zero.

What does it mean when factors in a factor matrix are "orthogonal”?

=uncorrelated so the correlation coefficient for any two factors would be equal to zero.

In what research design is autocorrelation a problem?

Repeated measures

refers to the correlation between measurements of the dependent variable when the dependent variable is repeatedly administered to the same subjects

What does a F-ratio of 1.0 suggest?

That the null hypothesis is true aka no effect

When should repeated measures ANOVA be used?

2+ correlated groups in a within-subjects design where exposed to 2 or more levels of the independent variable

What percentage of observations fall within two standard deviations of the mean in a normal distribution?

About 95%

What percentage of observations fall within three standard deviations of the mean in a normal distribution?

About 99%

What is a dependent variable (DV)?

The variable that is believed to be affected by the independent variable and is observed and measured.

What is a moderator variable?

A variable that affects the strength or direction of the relationship between independent and dependent variables.

Provide an example of a moderator variable.

Gender if a treatment is more effective for one gender than another.

What is a mediating variable?

A variable that explains or accounts for the relationship between independent and dependent variables.

Give an example of a mediating variable.

Self-efficacy beliefs may mediate the relationship between authoritative parenting and academic achievement.

what is sensitivity? how do you calculate it?

The proportion of all people who truly have a condition that are correctly identified as such

Aka, the “probability of detection”

-

TP / (TP + FN)

what is specificity? how do you calculate it?

ability of a test to detect the proportion of people who actually do not have the disease

- Aka, the “true negative rate”

TN / (TN + FP)

what is positive predictive value (PPV)? how do you calculate it?

The probability that following a positive test result, that indiv will truly have that specific disease

- considers prevalence

PPV = TP / (TP + FP)

what is negative predictive value (NPV)? how do you calculate it?

The probability that following negative test result, that indiv will truly not have that specific disease.

- considers prevalence

NPV = TN / (FN + TN)

What is interval recording in behavioral sampling?

A method that involves dividing a period of time into discrete intervals and recording whether a behavior occurs in each interval.

What is event sampling?

A method of behavioral sampling that records each occurrence of a behavior during a predefined or preselected event.

What distinguishes true experimental research from quasi-experimental research?

True experimental research allows for greater control and includes random assignment to groups, while quasi-experimental research permits less control.

What is random assignment?

The process of randomly assigning subjects to treatment groups, considered the hallmark of true experimental research.

How does random assignment differ from random selection?

Random assignment refers to assigning subjects to groups, while random selection refers to selecting subjects from the population.

What is probability sampling?

A sampling method where each element in the target population has a known chance of being selected for inclusion in the sample.

Name three methods of probability sampling.

Simple random sampling, stratified random sampling, and cluster sampling.

What is simple random sampling?

A method where every member of the population has an equal chance of being selected, and the selection of one person does not affect the next.

What is stratified random sampling?

A method that ensures representation of specific strata (characteristics) in the sample by dividing the population and randomly selecting from each stratum.

What are typical strata used in stratified random sampling?

Gender, age, education level, socioeconomic status (SES), and racial, ethnic, or cultural background.

What is cluster sampling?

A method that selects units (clusters) of individuals rather than individuals and includes all individuals in the selected units or randomly selects individuals from each unit.

What is multistage cluster sampling?

A technique where individuals are randomly selected from each cluster after the clusters have been selected.

Why is random sampling important?

It reduces the probabilities that a sample will be biased, especially when the sample size is large.

What is the significance of having at least two levels for each independent variable?

It allows for the examination of the effect of the independent variable on the dependent variable.

How do moderator and mediator variables differ in their roles?

Moderator variables affect the strength or direction of a relationship, while mediating variables explain the relationship between independent and dependent variables.

What is the hallmark of true experimental research?

Random assignment to treatment groups.

What is the main purpose of experimental research?

To conduct an empirical study to test hypotheses about the relationships between independent and dependent variables.

What is cluster sampling and when is it useful?

involves randomly selecting groups (clusters) and then including all or a random sample of individuals from those groups. It is useful when it is not possible to access the entire population of interest.

What are the three sources of variability an investigator should control in a research design?

(a) Variability in the dependent variable due to the independent variable, (b) variability due to extraneous variables, and (c) variability caused by random error.

How can an investigator maximize control over sources of variability?

By conducting a true experimental study, which provides the most control compared to quasi-experimental research.

What is systematic error and how does it relate to extraneous variables?

Systematic error is predictable error, and extraneous (confounding) variables are a source of systematic error that affects the relationship between independent and dependent variables.

What is the purpose of random assignment of subjects to treatment groups?

equalizes the effects of known and unknown extraneous variables, making it a powerful method of experimental control.

What is a primary shortcoming of holding extraneous variables constant?

It limits the generalizability of the research results.

How can matching subjects on an extraneous variable help in research?

ensures groups are equivalent regarding the extraneous variable and is useful when the sample size is too small for random assignment to guarantee equivalence.

What is the method of building an extraneous variable into a study?

It involves including the extraneous variable as an additional independent variable so that its effects can be statistically analyzed.

What does it mean to block subjects in a study?

means grouping subjects based on their status on an extraneous variable and then randomly assigning subjects within each block to treatment groups.

What is the significance of including two independent variables in a study?

allows for the analysis of their combined effects on the dependent variable.

What is the difference between true experimental research and quasi-experimental research?

True experimental research provides greater control over variability and uses random assignment, while quasi-experimental research lacks this level of control.

Why is maximizing variability in the dependent variable important?

It ensures that the observed variability is attributable to the independent variable rather than to error.

What is the role of extraneous variables in research?

can confound the results by affecting the relationship between the independent and dependent variables.

What is the effect of random error on research outcomes?

introduces variability that can obscure the true relationship between the independent and dependent variables.

How does matching subjects help when sample sizes are small?

It helps ensure that groups are equivalent with respect to the extraneous variable, compensating for the lack of random assignment.

What is the potential downside of selecting homogeneous subjects for a study?

It can limit the generalizability of the research results to a broader population.

What is the advantage of using blocking in a study?

It allows researchers to control for the effects of an extraneous variable while still conducting random assignment within blocks.

How can an investigator control variability due to extraneous variables?

By using methods such as random assignment, matching, holding variables constant, or blocking.

What is the relationship between independent and dependent variables in experimental research?

The independent variable is manipulated to observe its effect on the dependent variable.

What is the importance of controlling extraneous variables in research design?

helps ensure that the results accurately reflect the effects of the independent variable.

What is the significance of maximizing differences in independent variable levels?

It enhances the ability to detect effects on the dependent variable, providing clearer insights into the relationship.

What is an extraneous variable when included as an independent variable in a study?

It is also known as a moderator variable.

What statistical technique can be used to control an extraneous variable in research?

ANCOVA (analysis of covariance) can be used to remove variability in the dependent variable due to the extraneous variable.

What does ANCOVA do with regard to subjects' status on an extraneous variable?

It equalizes all subjects with regard to their status on that variable.

In what type of research is statistical control particularly useful?

In quasi-experimental research where subjects cannot be randomly assigned to treatment groups.

What is random error?

is unpredictable error, which includes sampling error and measurement error.

How can an investigator control for random error?

By ensuring subjects do not become fatigued, the experimental setting is free from distractions, and all measuring devices are reliable.

What does internal validity refer to in research?

The degree to which a study allows an investigator to conclude that observed variability in a dependent variable is due to the independent variable.

What is maturation in the context of internal validity?

A threat that occurs when a physical or psychological process affects subjects' status on the dependent variable due to the passage of time.

What is the history threat to internal validity?

It occurs when an external event affects subjects' performance on the dependent variable in a systematic way.

What does statistical regression refer to as a threat to internal validity?

It refers to the tendency of extreme scores to regress to the mean on retesting, particularly when subjects are selected based on extreme status.

How does selection threaten internal validity?

It threatens when groups differ at the beginning of the study due to the way subjects were assigned to groups.

What does external validity refer to?

The degree to which a study's results can be generalized to other people, settings, and conditions.

What is pretest sensitization in relation to external validity?

A threat that occurs when pretesting affects how subjects react to the treatment.

What is reactivity as a threat to external validity?

It occurs when subjects respond differently to a treatment because they know they are participating in a research study.

What is multiple treatment interference in external validity?

A threat that occurs when subjects receive more than one level of an independent variable.

What is counterbalancing used for in research?

To control multiple treatment interference by administering different levels of the independent variable to different groups in a different order.

What are between-groups designs in experimental research?

Designs that assess the effects of different levels of one or more independent variables by administering each level to a different group of subjects.

What are within-subjects designs in experimental research?

Designs in which each subject receives each level of the independent variable at different times.

What is a single-group time-series design?

A type of within-subjects design.

What are mixed designs in research?

a type of factorial design where at least one independent variable (IV) is a between-groups variable and one IV is a within-subjects variable.

What is a factorial design?

- Any study that includes 2+ IVs

- allows for the analysis of main and interaction effects

What is a main effect in factorial designs?

the effect of a single independent variable on the dependent variable, disregarding the effects of all other independent variables

What does interaction refer to in factorial designs?

when the effects of 2+ IVs considered together; occurs when the effects of an IV differ at different levels of another IV

What are single-subject designs?

include at least one baseline (A) and one treatment (B) phase, with multiple measurements of the dependent variable at regular intervals during each phase.

What is an AB design in single-subject research?

includes a single baseline phase (A) followed by a single treatment phase (B).

What is a reversal design in single-subject research?

include at least two baseline phases and one treatment phase, with the treatment being withdrawn during the second and subsequent baseline phases (e.g., ABA or ABAB design).

What is a multiple baseline design in single-subject research?

involves sequentially applying a treatment to different baselines, which can include different behaviors, settings, tasks, or subjects.

What are the four scales of measurement?

are nominal, ordinal, interval, and ratio, categorized from least to most mathematically sophisticated.

What is a nominal scale?

divides variables into unordered categories and yields frequency data, such as gender classification.

What is a limitation of the nominal scale?

The primary limitation is that the only mathematical operation that can be performed is counting the frequency of cases in each category.

What is an ordinal scale?

divides observations into categories and provides information on the order of those categories, allowing for comparisons of more or less.

What is an example of an ordinal scale?

Ranks and ratings on a Likert scale

What is a limitation of ordinal scores?

do not indicate how much difference exists between scores, only the order.

What is an interval scale?

has the properties of order and equal intervals between successive points on the measurement scale.

How does an interval scale differ from a nominal scale?

An interval scale provides not only categories but also order and equal distances between points, unlike a nominal scale which only categorizes.

What is an example of a variable measured on a nominal scale?

Gender, religion, or political affiliation

What type of data does an ordinal scale yield?

yield data that can be ranked but do not provide information on the magnitude of differences between ranks.

What is the significance of using a Likert scale in research?

provides ordinal data that reflects attitudes or opinions, allowing researchers to assess the degree of agreement or disagreement.

What are the characteristics of a ratio scale?

has all the properties of an interval scale, plus a true zero point, allowing for the comparison of absolute magnitudes.

In what scenario would a multiple baseline design be particularly useful?

useful when it is not possible or ethical to withdraw treatment from a participant.

What is the purpose of using factorial designs in research?

allow researchers to examine the effects of multiple independent variables simultaneously and their interactions.

What type of scale do standardized IQ tests represent?

An interval scale.

What mathematical operations can be performed on interval scores?

Addition and subtraction.

What is an example of an interval scale?

Temperature measured in Fahrenheit or Celsius.