QUESTIONS WITH CORRECT ANSWERS
GRADED A+
◍ quality improvement and evaluation.
Answer: Also requires institutional approval
◍ descriptive statistics for categorical variables.
Answer: frequencies percentages
◍ Dependent variable.
Answer: The outcome variable(s) In study of atiHTN drugs, the primary
outcome variable would be BPMany studies measure more than 1 (study can
include measurement of short term outcome of BP and long term outcome
of frequency of MI, stroke, age of death)
◍ selecting a statistical test.
Answer: 1. determine the level of measurement2. determine if you should
use parametric or nonparametric tests3. if there are groups, determine the
number of groups in the analysis (2, 3 or more)4. determine whether the
observations are independent or dependent5. determine what test should be
used
◍ convenience sample.
Answer: a form of nonprobability sample using respondents who are
convenient or readily accessible to the researcher—for example, employees,
friends, or relatives
◍ descriptive statistics for continuous variables.
Answer: measures of central location (mean, median, mode)measures of
spread (std, statistical range, epidemiological range, Q1, Q3, IQR)graphic
display: histogram, steam and leaf, box and whisker with turkey fences
,◍ measurement error.
Answer: an error that occurs when there is a difference between the
information desired by the researcher and the information provided by the
measurement processaccuracy of measurements
◍ scatterplot.
Answer: displays the relationship between ordinal and continuous variables
help determine whether there appears to be a relationship, direction of the
relationship, spread of observations, and shape of relationship (linear =
straight, nonlinear = curved)
◍ How to determine the presence or absence of normal distribution?.
Answer: By performing a statistical test for normality — Shapiro wilk test
Only done to determine whether you have a normal distribution so that you
can decide which type of descriptive statistics and statistics test should be
used
◍ when can parametric procedures be used?.
Answer: 1. interval or ratio level data (continuous) AND2. data are normally
distributed AND3. there are no outliers
◍ Which graphic display should be performed for nominal data?.
Answer: BAR CHART
◍ P value <0.5.
Answer: There IS a statistically significant difference between the
distribution of your variable and a normal distribution DATA IS NOT
NORMALLY DISTRIBUTED
◍ Pearson correlation coefficient.
Answer: parametric testassumes that both variables are measured at the
interval/ratio level, are normally distributed, no outliers
◍ Relative frequency.
Answer: The proportion of the total number of observations in each
category; calculated as a decimal and should be converted to percentage
, ◍ type 2 error.
Answer: occurs when:1. p value is not statistically significant (>.05)2. does
not reject the null hypothesis3. but there really is a significant difference
between the groups or relationship between the variables null hypothesis
was not rejected -- incorrectly concluded that there was not a difference
between groups when there really was oneFALSE NEGATIVE
◍ How many independent variables can there be in a study?.
Answer: Only one
◍ type 1 error.
Answer: occurs when:1. p value is statistically significant (<.05)2. rejects
the null hypothesis and accepts the alternative hypothesis3. but there is not
really a significant difference between the groups or relationship null
hypothesis was incorrectly rejected and research hypothesis was incorrectly
accepted -- researcher concluded there was a difference when there was
noneFALSE POSITIVE
◍ Quartiles.
Answer: Describe the values for which 25% (Q1), 50% (Q2), and 75% (Q3)
of the observations fall below
◍ Percentiles.
Answer: Describe the percentage of observations that fall below a certain
value
◍ univariate.
Answer: one variableanalysis should be performed on all variables
(continuous or categorical)
◍ What video will be discussed in class on 3/5?.
Answer: Why Everyone In Your Company Should Speak a Little English
◍ cross tabulation.
Answer: contingency table used to examine relationships between nominal
and ordinal variables with a small number of categorieschi square, fisher, or