206 exam
In a particular neighbourhood in Auckland there are 350 households in which English is usually spoken, 50 in which Te Reo (Māori) is usually spoken, and 100 in which the language is other than English or Te Reo. If a researcher approaches a house at random in the neighbourhood to conduct an interview, what is the chance that the usual language in that house will not be English?
150 / 500 = .3.
A researcher conducts a t test for dependent means in which it is predicted that there will be a decrease in anxiety from before to after a particular self-development program. The cutoff t needed is -1.833. The standard deviation of the distribution of means of change scores is 2.0, and the mean change score for the sample studied is an increase of 5.2.
5.2/-1.833 = -2.84
In a planned study in which an increase is predicted, the raw-score cut off on the comparison distribution corresponds to a Z score of -1 on the distribution of means representing the research hypothesis. What is the power of this study?
84%.
Two variables, A and B, have a strong linear correlation. What is the direction of causality?
Any of the above are possible.
Post-hoc comparisons
Examine any difference that show up after the study has been conducted.
When figuring the correlation coefficient, you first convert your scores to deviation scores. This is because with deviation scores,
High scores are positive numbers, and low scores are negative numbers.
What does the raw-score regression coefficient tell you?
How many units of change in the predicted value of the dependent variable you get from each unit change in the predictor variable.
What does it mean if a researcher said she rejected the null hypothesis at the .05 level?
It means that there was less than a 5% chance that she would have got such an extreme result by chance if the null hypothesis were true.
A researcher conducts a survey of job satisfaction among labourers and white-collar workers. (Job satisfaction is rated on a scale from 1, low satisfaction, to 10, high satisfaction). The findings are that labourers have a mean job satisfaction score of 4 and a variance of 1; white-collar workers have a mean job satisfaction rating of 7 and a variance of 10. Which statement best describes what these numbers mean?
Labourers are less satisfied than white-collar workers and most labourers are about equally dissatisfied while there is more variation in the satisfaction of white-collar workers.
When you compute the average of the cross-products of Z scores, it turns out that it is
Never more than +1.0 or less than -1.0
When the results of a study are not sufficiently extreme to reject the null hypothesis, what can the researcher conclude with reasonable confidence?
None of the above - the results are inconclusive.
Which statement is true for the scores 1,2,3,4,5,5,7,8,9,10,?
The mean is greater than the median.
What does it mean when SSTotal minus SSError equals zero?
This is the worst case: it means the prediction model has reduced zero error.
In which situation below would you use a t test for dependent means?
To compare the level of reading comprehension of students at the beginning of a speed-reading class to their level of reading comprehension at the end of the class.
Of the following rank-order tests, which applies to the same kind of research situation as a t test for independent means?
Wilcoxon rank-sum test
A population is usually NOT normally distributed if
a ceiling or floor effect exists.
Suppose that in general it turned out that eating a certain fruit before a test, and the more fruit eaten, the better people do on the test; but beyond a certain point, the more fruit eaten, the worse people do on the test. The relation between amount of fruit eaten and how well people do on the test is an example of
a curvilinear relationship.
For a one-way analysis of variance, if the null hypothesis is true, then
any difference among sample means reflects variance within the populations.
The REDUCTION in squared error computed in bivariate regression is the same as ______________________ in the analysis of variance.
between-group sum of squared deviations
In a chi-square test, the variables are
categorical (nominal).
Each of the following is part of conducting a t test for independent means, EXCEPT
difference scores are computed for each subject.
Psychologists use frequency tables, histograms, or frequency polygons to demonstrate
how the numbers they collect are distributed.
When testing the significance of the correlation coefficient, the null hypothesis is usually that in the population, the true correlation
is less than the actual correlation.
Type II errors concern scientists because
it could mean that a good theory or useful practical procedure is not used.
Once the null hypothesis is rejected as a result of an analysis of variance,
it is often useful to conduct some form of multiple comparison.
The normal curve is typically
kurtotic.
If you discover there is one extreme outlier in a sample of scores, the first thing you would do would be to
make sure the score was not recorded in error.
One way to determine the number of people expected to be in any one cell of a chi-square contingency table is to
multiply the number of people for the cell's column by its row proportion, figuring its row proportion as the number of people for the cell's row divided by the total number of people.
A statistics professor discovers that the mean score and the mode score for the latest class exam are the same. This would be possible for which of the following distribution shapes?
normal
The probability of NOT making a Type II error is called
power.
An independent variable can also be called a
predictor variable.
You figure a correlation in a situation in which only a limited range of possible values on one variable are included in the group studied. In this situation the correlation cannot be properly extended to apply to the entire range of values the variable might have among people in general. What is this called?
restriction in range
In a chi-square test of independence, the term "expected frequency" generally refers to
the distribution of people over categories on one variable expected if the distribution of people over categories on the other variable is completely unrelated to it.
When determining the pooled population variance estimate in a t test for independent means
the estimates based on each of the samples are averaged in such a way as to give more influence to the estimate based on more subjects.
The degrees of freedom for the chi-square test for independence is
the number of categories minus one.
The comparison distribution for a t test for independent means is a t distribution (as opposed to a normal curve) because
the population variance is estimated.
In a chi-square test for goodness of fit, the null hypothesis is that
the proportion of people over categories breaks down the same for the two populations.
Even if the research hypothesis is true, an experiment may fail to yield significant results because
the sample that happens to be selected from the population being studied may not be extreme enough to provide a clear case for rejecting the null hypothesis.
In the formula, " Y = a + (b1) (X1) + (b2) (X2) + e, " the "e" refers to
the sum of all variables that influence Y other than X1 and X2
In the formula, "SDifference2 = SM12 + SM22", the "SM12" refers to
the variance of the distribution of means for Population 1.
In an analysis of variance, if the null hypothesis is false, then
the variance within each sample is larger than if the null hypothesis were true.
In an analysis of variance, if the null hypothesis is true, then
there should be less variance among means of samples than if the null hypothesis were not true.
One important advantage of using effect sizes is that
they are a standardized score, making study to study comparisons easier.
You are carrying out an analysis of variance (sample sizes are equal). When you go from the estimated variance of the distribution of means to an estimated variance of the population of individual scores,
you multiply the estimated variance of the distribution of means by the number of scores in each sample.
