Statistics

The larger the value of the calculated F-ratio:

the less likely an observed difference is due to chance.

The variation observed between group means is referred to as:

between group variation

Which is NOT true of an Analysis of Variance?

All data must be ordinal level data.

Calculate the between-group degrees of freedom for an Analysis of Variance with three groups, each containing 20 respondents:

2

Calculate the within-group degrees of freedom for an Analysis of Variance with three groups, each containing 20 respondents:

57

In a one-way Analysis of Variance with three groups, the research hypothesis is:

at least two of the groups differ.

True of false? The ANOVA will let the research know if there is differences between the groups and if so the Tukey's HSD can be used to show were the differecnes are.

True

True or False?
In ANOVA if the F-ratio is larger than the critical value the groups are the same in the population (nonsignificant).

False

The variation found among raw scores in a particular group is referred to as:

within group variation.

Tukey's HSD test is used:

only after the Analysis of Variance has been computed and found to be significant.

researcher is interested to know if there is a difference in levels of religiosity between different income groups. The researcher administers a religiosity scale to a total of 15 people: 5 "low income" individuals, 5 "middle income" individuals, and 5 "h

Calculate the total sum of squares=74.4
Calculate the within-group sum of squares.=27.6
Calculate the between-group sum of squares.=46.8
Within-groups degree of freedom=12

Degrees of freedom for a two-way Chi-square test depends on the number of rows and columns.

True

A Chi-square tests depends on nominal and ordinal levels of measurement.

True

The frequencies proposed under the terms of the null hypothesis are the:

expected frequencies.

You take a sample and want to compare the results to the population from which it was drawn. The independent variable is "race" and the dependent variable is a yes/no response to whether they favor the death penalty. What test would you use to see if your

a chi-square test

PARAMETRIC tests require:

a normal distribution.

The median test determines:

the likelihood that the samples were drawn from populations with equal medians.

The term"power" refers to:

the probability of correctly rejecting a false null hypothesis.

As the observed frequencies get closer to the expected frequencies, the value of the chi-square statistic:

decreases

The degrees of freedom for a one-way chi-square statistic are:

K-1

Which of the following is true of the one-way chi-square test?

The more the observed frequencies deviate from the expected frequencies, the greater the value of chi-square

A Chi-square test is a parametric test of significance.

False

Which of the following is NOT a requirement of the median test?

Nominal Data

The strength of a correlation is indicated by:

size

The degrees of freedom for a two-way chi-square statistic are:

(r-1) (c-1).

Degrees of freedom for a Chi-square test depends on the sample size.

False

NONPARAMETRIC tests require:

None of the above are required.

If a chi-square expected frequency is less than 10, one should:

use Yates' correction.

A chi-square test should never be used when:

there is an observed frequency less than 5.

Expected frequencies represent:

the frequencies one would expect if the null hypothesis were true.

The power of a test refers to:

the probability of rejecting the null hypothesis when it is actually false

The direction of the correlation is indicated by:

+ or -

A negative correlation between variables X and Y implies:

high scores on X are associated with low scores on Y.

A positive correlation between variables X and Y implies:

high scores on X are associated with high scores on Y.

Choose the weakest correlation: -.30 or +.12

+.12

Which statement is NOT true of correlation coefficients?

correlations provide statements of causation.

If a correlation between variables A and B is found to be equal to 1.35:

we have miscalculated.

If we correlated people's height with their shoe size, the correlation would probably be:

positive

If variable X increases as variable Y increases, the correlation is said to be:

positive

If variable X increases as variable Y decreases, the correlation is said to be:

negative

If variable X increases as variable Y remains constant, the correlation is said to be:

0 zero

The correlation between two variables that are totally unrelated would be:

0 zero

An important first step in assessing the relationship of two interval level variables is:

look at scatterplot

A strong curvilinear relationship between two variables might yield a Pearson's r that is

weak or close to 0

If the researcher claims a significant difference between groups, when in fact none exists:

a Type 1 error is made.

As the alpha level INCREASES:

the probability of making a Type 1 error increases.

The critical value for assessing the difference between two proportions at the 0.05 level is 1.96.

true

The larger the value of our obtained t:

the less probable that our results are due to chance alone.

which statement is true of alpha?

Conventionally researchers set alpha = .05.

The null hypothesis in a two-tailed, independent t-test of means would be:

mean1 equals mean2

When we accept the null hypothesis, we:

conclude that sampling error is responsible for our obtained difference.

If we accept a null hypothesis that is in fact false:

Type 2 error

the variance of the sampling distribution of differences between means is equal to that of the population variance.

False

If the calculated t-value is larger than the critical value of t we reject the null hypothesis.

True

if the null hypothesis is true, there is no difference between population means.

True

Alpha is the level of probability at which the null hypothesis can be rejected with confidence.

True

If our obtained t-value is greater than our critical t-value, we:

reject the null hypothesis as false.

Which of the following is not required for testing the differences between sample means

population mean

f we conduct a two-sample independent t-test of significance on a sample containing 50 subjects PER GROUP(n=100), our degrees of freedom are:

98

If the researcher fails to find a significant difference, when in fact one exists in the population:

a Type 2 error has been made.

We have collected SAT scores from 26 students Before and After completing an SAT review course. The degrees of freedom for our t-test are:

25

Compared to the .05 level of significance, the .01 level of significance:

means a lower probability of the obtained difference being a result of sampling error.

The t-distribution more closely approximates the distribution of the normal curve when:

the degrees of freedom increase thus a larger sample size. .