WGU C784 -APPLIED HEALTHCARE STATISTICS
OBJECTIVE ASSESSMENT
When should the IQR be used? - Only when the median is used as a measure of
center.
When should an outlier be kept in the data? - -when the outlier is produced by the same
sort of physical or biological process as the rest of the data.
-the outlier is expected to eventually occur again
five number summary - min, Q1, median, Q3, max
standard deviation - Quantifies the spread of a distribution by measuring how far the
observations are from their mean. ( gives the average distance between a data point
and the mean (xbar)
how to find standard deviation? - -find the mean
-find deviations from the mean (xbar - 2,3,4,5 etc)
-square each deviation
-add all squared deviations then divide them by n - 1 (sample size minus 1)
-find square root of new total, this is the standard deviation
Variance of data - The average of the squared deviations
When to use x-bar and SD as numerical summary? - Only for reasonably symmetric
distributions with no outliers
When to use the median and IQR as a summary? - All other cases that are not
symmetric
How many observations fall within 1 SD of the mean? (Normal distribution) - 68%
How many observations fall within 2 SD of the mean? (Normal distribution) - 95%
How many observations fall within 3 SD of the mean? (Normal distribution) - 99.7%
explanatory variable - Also called the independent variable.
The variable that claims to explain, predict, or affect the response. Denoted by X.
response variable - Also known as the dependent. The outcome of the study. Denoted
by Y.
Case C -> Q - -data displayed with side by side box plots
, -numerical summaries: descriptive statistics
Case C -> C - Use two way table to simplify and summarize data. Use conditional
percentages in two way table. Find conditional percentages by dividing the cell number
by total.
Case Q -> Q - Use a scatterplot
Where should the explanatory variable be placed? and the response? - The explanatory
should always be put on the horizontal X-axis. The response should always be on the
vertical Y-axis.
Correlation Coefficient r - The numerical measure used to asses the strength of a linear
relationship.
What are the four points of correlation coefficient r? - -The value of r ranges from -1 - 1
-Values of r that are close to 0 either positive or negative indicate a weak relationship.
-Values close to -1 or 1 indicate a strong relationship
-r is unit-less. Changing the type of units the response variable will not change the r.
How and do outliers affect (r)? - Outliers affect correlation coefficient r greatly. An outlier
going with the general form will strengthen (r) and if it is not it will decrease.
Regression Lines - A technique that specifies the dependence of the response variable
on the explanatory variable.
Linear Regression - When the dependence of the response and explanatory is linear.
How do you find the linear regression? - We use the least squares criterion. Among all
the lines that look good on your data, choose the line that has the smallest sum of
squared vertical deviations. Y = a+bX & B = r(Sy/Sx)
Extrapolation - The prediction of variables that are not in the data. This is not a reliable
source and should generally be avoided.
nominal variables - Categorical variables that there is no natural order among the
categories
ordinal variable - Categorical variables where there is natural order among the
categories (low-high)
interval - A measurement which makes sense to talk about the difference between
values but not the ratio between values
OBJECTIVE ASSESSMENT
When should the IQR be used? - Only when the median is used as a measure of
center.
When should an outlier be kept in the data? - -when the outlier is produced by the same
sort of physical or biological process as the rest of the data.
-the outlier is expected to eventually occur again
five number summary - min, Q1, median, Q3, max
standard deviation - Quantifies the spread of a distribution by measuring how far the
observations are from their mean. ( gives the average distance between a data point
and the mean (xbar)
how to find standard deviation? - -find the mean
-find deviations from the mean (xbar - 2,3,4,5 etc)
-square each deviation
-add all squared deviations then divide them by n - 1 (sample size minus 1)
-find square root of new total, this is the standard deviation
Variance of data - The average of the squared deviations
When to use x-bar and SD as numerical summary? - Only for reasonably symmetric
distributions with no outliers
When to use the median and IQR as a summary? - All other cases that are not
symmetric
How many observations fall within 1 SD of the mean? (Normal distribution) - 68%
How many observations fall within 2 SD of the mean? (Normal distribution) - 95%
How many observations fall within 3 SD of the mean? (Normal distribution) - 99.7%
explanatory variable - Also called the independent variable.
The variable that claims to explain, predict, or affect the response. Denoted by X.
response variable - Also known as the dependent. The outcome of the study. Denoted
by Y.
Case C -> Q - -data displayed with side by side box plots
, -numerical summaries: descriptive statistics
Case C -> C - Use two way table to simplify and summarize data. Use conditional
percentages in two way table. Find conditional percentages by dividing the cell number
by total.
Case Q -> Q - Use a scatterplot
Where should the explanatory variable be placed? and the response? - The explanatory
should always be put on the horizontal X-axis. The response should always be on the
vertical Y-axis.
Correlation Coefficient r - The numerical measure used to asses the strength of a linear
relationship.
What are the four points of correlation coefficient r? - -The value of r ranges from -1 - 1
-Values of r that are close to 0 either positive or negative indicate a weak relationship.
-Values close to -1 or 1 indicate a strong relationship
-r is unit-less. Changing the type of units the response variable will not change the r.
How and do outliers affect (r)? - Outliers affect correlation coefficient r greatly. An outlier
going with the general form will strengthen (r) and if it is not it will decrease.
Regression Lines - A technique that specifies the dependence of the response variable
on the explanatory variable.
Linear Regression - When the dependence of the response and explanatory is linear.
How do you find the linear regression? - We use the least squares criterion. Among all
the lines that look good on your data, choose the line that has the smallest sum of
squared vertical deviations. Y = a+bX & B = r(Sy/Sx)
Extrapolation - The prediction of variables that are not in the data. This is not a reliable
source and should generally be avoided.
nominal variables - Categorical variables that there is no natural order among the
categories
ordinal variable - Categorical variables where there is natural order among the
categories (low-high)
interval - A measurement which makes sense to talk about the difference between
values but not the ratio between values
