Bcacp Statistics Study Guide Exam Questions And Answers 2024.

Ho = The Null Hypothesis Accepted when no statistical difference was detected between treatment groups; p-value 0.05 Ha = The Alternative Hypothesis Opposite of the null hypothesis; accepted when there is a statistical significance between the treatment and placebo groups; p-value 0.05 Independent Variables Variables set by the researcher; can be controlled Dependent Variables Variables that will DEPEND on the independent variables; can not be controlled P-value (alpha) and statistical point Likelihood of a Type 1 error o.05 by convention Type 1 Error Detecting a difference when in reality one doesn't actually exist (a false positive) Beta Probability of making a Type 2 error Type 2 Error The percentage or chance of NOT detecting a difference when once actually exists (a false negative) Power of a study How well a study is made to detect a significant difference Power = ? 1- Beta What is considered a good benchmark power for a study? 80% The ______beta is, the _______ power your study will have higher; less OR lower; more Categorical Variables Nominal and Ordinal They cannot have any value between two points Nominal Variables Variables that have two or more groups but there is no definite order to these variables - can you answer "yes or no" or can you "name" them into a category What are a few classic examples of nominal variables? Male/female Smoker/non-smoker Patients with diabetes vs those without Eye color Ordinal Variables Variables that have an order, but they do not have a clear and easily interpreted difference between each value - you cannot precisely define the differences between values but have an order What are some classic examples of ordinal variables? Pain scale of 1-10 Surveys Quantitative Variables Continuous and Discrete Continuous Variables Ratio Scale and Interval Scale Variables that have a distinct, measurable distance between each value - you can have an infinite amount of values between two points Ratio variables Have set and identifiable differences between each point, and have an absolute zero What are some classic examples of a ratio variable? Length Weight Temperature in Kelvin Interval variables Have set and identifiable differences between each point, but has NO absolute zero What is a classic example of interval variable? Temperature in Celsius (Because you can have negative values which doesn't allow for standard arithmetic) Discrete Variables A countable type of variable that will always be a WHOLE number How does continuous differ from discrete variables? Continuous you can have fractions of a number whereas discrete you cannot - example #CHF exacerbation or Accuchecks are discrete because they cannot be fractioned numbers Parametric Data Refers to data that is associated with a normal distribution What variables use parametric data? Continuous variables like ratio and interval Non-Parametric Data Refers to data that is NOT associated with a normal distribution What variables use non-parametric data? Nominal or ordinal data What is normal distribution? Dealing with continuous variables gives you a classic bell shaped curve of data points which you will test with parametric tests Measures of Central Tendency How to assess the "middle" or "central" portion of the data Mean Median Mode Mean The average of all the values added up together and divided by the total number of values Standard Error of Mean Estimated interval where the mean actually lies based upon our sample taken Example: 10+/-2 means the true value will lie between 8 and 12 A ______ standard deviation or a larger sample size will make this range smaller smaller Median The middle number of all the data Mode The value that occurs most frequently What is assumed about the mean, median, and mode in a normal distribution? All these values are equivalent (i.e. the top of the bell) Negatively skewed data The long tail of the graph points toward 0 or is headed toward "negative" territory Positively skewed data The long tail of the graph is heading away from 0 with the tall peak on the left Standard Deviation A measure of the range or dispersion of the data i.e. how spread out the data is A _______ standard deviation the ___________ the bell curve will be Larger & flatter OR smaller & taller In a normal distribution, one SD away from the mean in each direction contains about how much of the data? 68% of the data In a normal distribution, two SD away from the mean in each direction contains about how much of the data? 95% of the data, this is where alpha is derived from Statistical test used most of the time for Nominal data Chi-squared test Common statistical tests used for Ordinal data Mann-Whitney u test or Wilcoxon signed Ranks test Statistical test used to compare two paired samples with nominal data Sign test Statistical test used to compare two unpaired samples with nominal data Chi-squared or Fisher-s exact test Statistical test used to compare three or more unmatched samples with nominal data Chi-squared test Statistical test used to compare two paired samples with ordinal data Sign test or Wilcoxon test Statistical test used to compare two unpaired samples with ordinal data Mann-Whitney test Statistical test used to compare three or more unmatched samples with ordinal data Kruskal-Wallis test Statistical test common with continuous data used for 1 independent variable with two or less groups T test (sometimes student t test /one sample t test) Statistical test used to compare two paired samples with continuous data Paired t test Statistical test used to compare two unpaired samples with continuous data