ARM final exam

. Which of the following statements is/are true?

Commonality is the proportion of variance in a variable explained by the extracted factors

The multiple dichotomy method of coding is used with:

Closed questions where more than one response option may be selected

. Which of the following are assumptions of survey research

*Respondents are willing to give the required information to researchers *Respondents have the information the researcher requires

Which of the following is an example of a parametric test?

*analysis of variance * one-sample t-test * Pearson correlation

Using a two-way between-subjects ANOVA with Factor A (Sex: male, female) and Factor B (Type of employment: blue collar, white collar), a researcher found that scores for men and women significantly varied across the levels of the second factor (type of employment). In this study, the researcher found a significant.

2 × 2 interaction

A researcher reports the following equation for a best-fitting straight line to a set of data points: . Which value is the y-intercept?

3.24

A researcher records the frequency of participants selecting one of four new slogans for an advertising campaign. If it was expected that among 200 people polled there would be no preference for any one slogan, then what was the expected frequency for each slogan.

50 people

Singularity in multiple regression refers to

A situation where one predictor variable is entirely predicted by other predictor variables

Here are selected statistics from a hypothetical Principal Components Analysis output (there is a graph for this one)

ECSTATIC and SATISFIED

A test with a systematic error close to zero, that minimises random error, will:

Have a high true score, and high reliability and validity

If you want to ensure that different age groups are included in your probability sample in the proportions in which they occur in the population, you would need to use:

Stratified sampling

. A group of students take an Advanced Research Methods test just before the end of semester and one when they return to school at the beginning of the next semester. The scores on both tests are examined for a correlation. This is an example of measuring

Test-retest reliability

What is the problem with the following data for computing a correlation? (there is a graph for this one)

The correlation coefficient will equal 1.0 because it violates the assumption of normality

The KMO statistic indicates:

Whether the correlation matrix is suitable for factor analysis or PCA

. A researcher conducts a within-subjects design and notes that scores increased at each level of the factor for all participants. Based on this description, does this test have high power to detect the effect?

Yes, because participant responding was consistent

What makes multiple regression different from correlation analysis?

You can include more than one predictor variable. You can control for shared variance in the predictors.

A researcher has participants complete a survey regarding their level of restraint when eating. Participants were assigned to groups based on their current health status (lean, overweight, obese) and gender (male, female). What type of factorial design is most appropriate for this study?

a between-subjects factorial design

The Bonferroni procedure is well adapted for which of the following research designs in which the same participants are observed in each group?

a within-subjects design*

The normality assumption states that the population of X and Y scores form a bivariate ("two variable") normal distribution, such that

a. the population of X and Y scores are normally distributed b. for each X score, the distribution of Y scores is normally distributed c. for each Y score, the distribution of X scores is normally distributed

What is the problem with the following data for computing a correlation? Factor 1 Factor 2 3 3 3 3 3 3 3 3 3 3

b. The correlation coefficient will equal 1.0 because it violates the assumption of normality

The assumption that there is an equal variance or scatter of data points dispersed along the regression line is referred to as

c. homoscedasticity

The interpretation of a 2 × 2 chi-square test for independence is like what other type of statistical test?

d. two-independent sample t-test

Ratio

data is genuine count data with an absolute zero. Income is an example.

Categorical

data is where people are classified, in a mutually exclusive way into one group or another. Being pregnant or not is an example.

Ordinal

data is where responses have a logical or order but there is no requirement that the distance between each point is exactly the same. Likert data with a scale of 1-5 is an example.

The most important psychometric properties of a scale:

is the validity

Construct/criterion validity

is where you examine if a measure is correlated with older/similar measures of the same contruct

Discriminant validity

is where you examine what the measure is not related to.

Convergent validity

is where you expect a measure to converge with results of related measures on some other measures.

Predictive validity

is where you expect a measure to predict some real-world outcome.

The appropriate correlation coefficient for measuring the direction and strength of the linear relationship between one continuous and one dichotomous variable is

the point-biserial correlation coefficient