Chapter 16: Data Analysis: Descriptive and Inferential Statistics

What percentage of scores fall within 1 standard deviation (SD) of the mean? A. 0% B. 34% C. 68% D. 95%

C. 68% 68% of all scores fall with a -1 SD and +1 SD from the mean. 95% of all scores fall within 2 SD of the mean.

Of the measures of central tendency, the most stable is considered to be what? A. Median B. Ratio C. Mean D. Interval

C. Mean The most stable measure of central tendency is the mean. The median is not considered to be a stable measure of central tendency. Ratio is a level of measurement rather than a measure of central tendency. Interval is a level of measurement rather than a measure of central tendency.

What is the best way to describe the distribution of the ages of all people who died in 2013? A. Normal distribution B. Positively skewed C. Negatively skewed D. Symmetrically distributed

C. Negatively skewed Age of death in the United States has a negative skew because most deaths occur at older ages. Most people die at ages later than the mean age of the population. Most deaths occur after childhood.

What are the levels of measurement, in ascending order? A. Interval, ordinal, nominal, ratio B. Ordinal, nominal, interval, ratio C. Nominal, ordinal, interval, ratio D. Nominal, ordinal, ratio, interval

C. Nominal, ordinal, interval, ratio The levels of measurement in ascending order are nominal, ordinal, interval, and ratio measurements.

A statistic is a characteristic of what? A. Normal distribution B. SD C. Sample D. Hypothesis

C. Sample The range is most affected by high and low scores. Using normal distribution in the analogy is not logical. An SD is a statistic. A statistic is not a characteristic of an SD. Using hypothesis in the analogy is not logical.

Which level of measurement is ranked variables or events with unequal intervals between numbers on the scale? A. Ratio B. Nominal C. Interval D. Ordinal

D. Ordinal Ordinal data are rankings of variables or events with unequal intervals. Ratio data are rankings of variables on scales with equal intervals. Nominal data are not rankings. They classify variables into categories. Interval data have equal intervals between the numbers.

A frequency distribution is used to do what? A. Determine the mean of the distribution. B. Determine the symmetry of the distribution. C. Demonstrate the measure of variability. D.Organize data by the frequency of response.

D. Organize data by the frequency of response. A frequency distribution groups the number of times each response is given. The mean is calculated by averaging the responses. It is the sampling distribution that determines the symmetry of the distribution. The measure of variability cannot be determined from the frequency distribution.

Which measure of variability is the most unstable? A. Mode B. Standard deviation C. Percentile D. Range

D. Range The range is most affected by high and low scores. The mode is not a measure of variability. The SD is a more stable measure of variability. A percentile is a more stable measure of variability.

Which level of measurement focuses on rank ordering with equal intervals and absolute zero? A. Nominal B. Ordinal C. Interval D. Ratio

D. Ratio Ratio measurement focuses on rank ordering with equal intervals and absolute zero. Nominal measurement focuses on classification. Ordinal measurement focuses on relative rankings. Interval measurement focuses on rank ordering with equal intervals.

A percentile of 90 is exceeded by what percentage of the scores.? A. 10% B. 45% C. 66% D. 90%

A. 10% The percentile represents the percentage of cases a given score exceeds. In this case, a percentile of 90 exceeds all but 10% of the scores.

When data are skewed and the researcher is interested in the "typical" score, what is the most appropriate measure of central tendency for the nurse researcher to use? A. Median B. Mode C. Mean D. Normal curve

A. Median The median is not sensitive to extremes in high and low scores. The mode is the middle, not typical, score. The mean is the average of scores, not the typical score. The normal curve is not a measure of central tendency.

Which level of measurement focuses on classification? A. Nominal B. Ordinal C. Interval D. Ratio

A. Nominal Nominal measurement focuses on classification. Ordinal measurement focuses on relative rankings. Interval measurement focuses on rank ordering with equal intervals. Ratio measurement focuses on rank ordering with equal intervals and absolute zero.

What occurs when differences between scores are too large to be caused by chance? A. Null hypothesis is rejected. B. Scientific hypothesis is proven to be true. C. Intervention did not make a difference. D. Scientific hypothesis is rejected.

A. Null hypothesis is rejected. The null hypothesis (the hypothesis that there are no differences) is rejected when there are differences between variables. It is impossible to prove that a scientific hypothesis is true. The differences are assumed to be caused by the intervention when they are too large to be caused by chance. With large differences, there is support for the scientific hypothesis.

Which level of measurement has a scale with an absolute zero? A. Ratio B. Nominal C. Interval D. Ordinal

A. Ratio Ratio data are rankings of variables on scales with an absolute zero. Nominal data are not rankings. They classify variables into categories. Interval data have an arbitrary zero. Ordinal data are rankings of variables or events with no absolute zero.

Which statements are applicable to type II errors? A. They occur when the sample size is too small. B. Mean rejects the null hypothesis when it is actually true. C. Mean accepts the null hypothesis when it is true. D. It can be prevented by setting the level of significance before the study begins.

A. They occur when the sample size is too small. A type II error is acceptance of a null hypothesis that is actually false, which occurs when the sample is too small, thereby limiting the opportunity to measure the treatment effect. The researcher can control the risk of type I errors by setting the level of significance before the study begins.

When data are positively skewed, where will the mean be located? A. To the right of the median B. To the left of the median C. The same as the mode D. Closer to zero than the mode

A. To the right of the median With skew, the data are distributed asymmetrically and have the peak off center. A positive skew means that the tail of the curve is to the left and the mean is to the right of the middle (median) score. A negative, not positive, skew means that the mean is to the left of the median. The mean and the mode are not the same when data are skewed. A negative, not positive, skew means that the mean is closer to zero.

When are parametric tests of significance used? A. Measurement is on a nominal scale. B. At least one population parameter is estimated. C. The sample is small. D. Normal distribution cannot be assumed.

B. At least one population parameter is estimated. Parametric tests require measurement on at least an interval scale. Parametric tests require estimation of at least one population parameter. Non-parametric tests are used when the sample size is small. Parametric tests assume normal distribution of the data.

The nurse researcher is interested in performing parametric tests of association on more than two variables. The most appropriate tests to use would consist of what? A. Chi-square B. Multiple regression C. Kendall's tau D. Spearman rho

B. Multiple regression Chi-square is a nonparametric statistic used to determine whether the frequency of each category is different from what would be expected by chance. Multiple regression should be used when the researcher needs to run parametric tests of association on more than two variables. Kendall's tau should be used with nonparametric tests of association that are run on only two variables. Spearman rho should be used with nonparametric tests of association that are run on only two variables.

Which level of measurement focuses on relative rankings? A. Nominal B. Ordinal C. Interval D. Ratio

B. Ordinal Ordinal measurement focuses on relative rankings. Nominal measurement focuses on classification. Interval measurement focuses on rank ordering with equal intervals. Ratio measurement focuses on rank ordering with equal intervals and absolute zero.