8 Understanding Statistics
Interval measurement
-Amount of an attribute, equal distance between each number in terms of amount -No absolute 0 value -Used in parametric tests
Nominal data
-Categories -Reported by a percentage -Can't treat nominal data mathematically -Can be grouping data -Logistic regression results in odds ratios -Used in nonparametric tests -Can't be put in order
Parametric tests
-Every statistical test has assumptions that must be met -Parametric tests have more stringent assumptions -Two of the most critical assumptions that parametric tests have are normal distribution of the data and level of measurement -Parametric tests are more powerful
Ratio
-Sames as interval except there is an absolute 0 -Examples: morphine in mg, length in in, temperature in K -Can make comparative statements such as "20 lbs is twice as heavy as 10 lbs" -Used in parametric tests
Ordinal measurement
-Uses number to designate the ordering of an attribute -Not an equivalent distance between each number -Examples: SES, academic rank, psychological inventories -Used in nonparametric test -Data do not have units
Which of the following variables is not measured at the nominal level? A. age B. race C. gender D. political affiliation
A
Nonparametric tests
A family of statistical procedures that do not rely on the restrictive assumptions of parametric tests. In particular, they do not assume that the sampling distribution is normally distributed.
Which of the following is an example of a discrete variable? A. Annual income B. Number of emergency department visits in one year C. Cholesterol level D. Score on the Mini-Mental State Exam
B
In the following hypothesis what is the independent variable? The annual income of MBA graduates is higher than the annual income of MSN graduates. A. There is no independent variable, the hypothesis is simply asking for a description of the variables. B. Income is the IV C. The grouping variable or IV is MBA versus MSN D. None of the above
C
What is the DV in the following research hypothesis? Women age 65 and older who take recommended calcium supplements will have greater bone density than women age 65 and older who do not take recommended calcium supplements. A. Women who take calcium supplements and those who do not B. The recommended dose of calcium C. Bone density D. None of the above
C
Which of the following is measured at the ordinal level? A. White blood cell count measured using laboratory methods B. Race measured using categories approved by the US census department C. Educational level measured by grade school, high school or greater than high school D. Temperature measured using the Kelvin scale.
C
In analyzing a research question or hypothesis it is important to ask which of the following questions? A. What is the research question or hypothesis asking? B. What is the level of measurement of the IV and the DV? C. Are assumptions of the statistical test met, particularly normal distribution of the DV? D. All are correct.
D
To use inferential parametric statistics which of the following must be present in the data collected in the study? A. normal distribution of the dependent variable B. the dependent variable is measures at the interval/ratio level or is interval like in nature C. the independent variable must have 3 groups D. both A and B are correct
D
For nominal data, parametric statistical tests are most frequently used. True False
False
In general, in statistics interval and ratio data can be analyzed in the same fashion. True False
True
Parametric statistics are generally considered more powerful than non-parametric statistics because they are more likely to find statistical significance if the alternative hypothesis is true ( for example, a difference when there is a difference). True False
True