Choose the alternative that best completes the stem of each question.

1		Exploratory data analysis should be used to
			A)	help you search for patterns in your data.
			B)	spot serious defects in your data that may warrant taking corrective action.
			C)	help determine whether assumptions of the inferential tests you intend to use may have been violated.
			D)	all of the above
2		___________________ are used to represent category values (e.g., gender) as values.
			A)	Unstacked formats
			B)	Dummy codes
			C)	Stacked formats
			D)	Codes
3		A drawback to organizing data into groups is that
			A)	the average performance of the group may not accurately represent the individual scores in the group.
			B)	data organized in group form cannot be analyzed statistically.
			C)	the average performance of the group usually overrepresents the individual scores in the group.
			D)	both a and b
4		The horizontal axis of a graph is called the ___________, whereas the vertical axis is called the __________.
			A)	ordinate; abscissa
			B)	y-axis; x-axis
			C)	x-axis; y-axis
			D)	linear; curvilinear
5		A bar graph is the best graph to use when
			A)	your dependent variable was measured on at least a ratio scale.
			B)	your independent variable is categorical.
			C)	your independent and dependent variables are both continuous.
			D)	you want to show ordered trends in your data.
6		To show a functional relationship between your independent and dependent variables, the graph of choice would be a
			A)	line graph.
			B)	histogram.
			C)	pie chart.
			D)	scatterplot.
7		A curve showing a functional relationship that starts off flat, becomes progressively steeper, and shows a single direction of change is
			A)	negatively accelerated.
			B)	monotonic.
			C)	positively accelerated.
			D)	both b and c
8		According to the text, graphing is important when
			A)	deciding on an appropriate research strategy.
			B)	your dependent variable was measured on at least a ratio scale.
			C)	you cannot possibly describe your data in any other way.
			D)	you are deciding on the appropriate descriptive statistic to apply to your data.
9		A ________ distribution has most scores collected about the center and is symmetrical about its midpoint.
			A)	functional
			B)	normal
			C)	monotonic
			D)	bimodal
10		A functional graph that shows a uniformly increasing or decreasing functional relationship is said to be
			A)	monotonic.
			B)	negatively skewed.
			C)	normal.
			D)	positively skewed.
11		An advantage of the stemplot for displaying a frequency distribution is that
			A)	you can recover the actual score values from the plot.
			B)	it automatically gives you the mean and standard deviation of the distribution.
			C)	it allows you to use any class width you wish.
			D)	all of the above
12		A ___________ is used to provide a single score representing an entire distribution of scores.
			A)	measure of correlation
			B)	measure of center
			C)	measure of spread
			D)	summary statistic
13		The middle score in an ordered distribution is the
			A)	mode.
			B)	mean.
			C)	median.
			D)	interquartile mean.
14		The range is the ___________ measure of spread, but it is also ___________
			A)	most difficult to compute; the most informative
			B)	easiest to compute; the least informative
			C)	most preferred; the most difficult to compute
			D)	none of the above
15		Dr. Jones measures how responsive a mother is to her infant by rating her behavior on a 10-point scale. He also obtains a measure of the strength of the infant’s attachment to the mother. He finds that as maternal responsiveness increases so does the strength of the attachment. This is an example of a(n)
			A)	negative correlation.
			B)	inverse relationship.
			C)	nonmonotonic relationship.
			D)	positive correlation.
16		In which of the following situations would you not want to use a Pearson correlation coefficient?
			A)	when the relationship between variables is nonlinear
			B)	when both of your variables are measured on at least an interval scale
			C)	when the variances of your distributions are very similar
			D)	all of the above
17		If one of your two measures is you wish to correlate is measured on a nominal scale and the other on an interval scale, ________ would be the most appropriate measure of correlation.
			A)	the Phi Coefficient
			B)	linear regression
			C)	the Point-Bisereal Correlation
			D)	the Spearman Rank Order Correlation
18		The Spearman Rank Order Correlation is used when
			A)	your data are scaled on an ordinal scale.
			B)	your data are scaled on an interval scale.
			C)	one measure is scaled on a nominal scale and the other on an ordinal scale.
			D)	one measure is scaled on a in interval scale and the other on an ordinal scale
19		_________ is used to estimate values of a variable based on knowledge of others.
			A)	The Spearman Rank Order Correlation
			B)	Linear regression
			C)	The Point-Bisereal Correlation
			D)	The Pearson Correlation Coefficient
20		The best fitting straight line describing a set of data points on a scattergram is known as the
			A)	residual line.
			B)	bivariate regression line.
			C)	Phi line
			D)	least squares regression line.
21		The regression weight (b) in the regression equation provides
			A)	the slope of the regression line.
			B)	the y-intercept.
			C)	an estimate of error in the regression equation.
			D)	a constant added to each predicted score.
22		Because raw score regression weights are difficult to interpret, you should use the _________ to interpret the results from a regression analysis.
			A)	y-intercept
			B)	standardized regression weights
			C)	Pearson correlation coefficients
			D)	constant
23		The difference between the value of a predicted and obtained score in a regression analysis is known as
			A)	the coefficient of nondetermination.
			B)	the Phi coefficient.
			C)	a residual.
			D)	a remainder.
24		The estimate of the amount of error in prediction of Y from X in a regression analysis is known as the
			A)	standard deviation.
			B)	coefficient of nondetermination.
			C)	averaged residual.
			D)	standard error of estimate.
25		Squaring a correlation coefficient yields the
			A)	coefficient of determination.
			B)	coefficient of nondetermination.
			C)	standardized regression weight.
			D)	standard error of estimate.