02.08 Common Stat tests

Today we’re going to be talking about a few common stats tests.

Three common statistical tests are the t-test, linear regression analysis and chi squared test.

Three main types of t-Tests. one Sample t-Test: compare mean of one group to a known mean does a drug change life expectancy? Using a control in a study small study and then comparing to a larger population. Paired Sample t-Test: Compare means from the same group at different intervals. Weight loss surgery costs at different hospitals Weight measurements using two different scales. Independent Sample t-test compares the means of two different groups. Most common t-test compares means of two data sets

  Drug effectiveness in males vs female.

Linear regression is when we draw a line of best fit between the plotted data points in hopes of r finding a linear relationship between the two variables.  Her had a positive linear relationship shown with the line of regression created for the reaction temp vs the yield produced from a chemical reaction. The closer to zero the r2 value is the more the variables are uncorrelated. a value trending towards 1.0 shows a positive correlation and a value trending towards -1.0

The chi square test allows us to test the goodness of fit between what was expected and what was observed. here is an experimental example with fruit flies that are in a contained environment. One end has regular water and the other sugar water. If there is no dif then one would expect ½ the flies at one end and ½ at the other.  So if there were 10 fruit flies 5 at one end and 5 at the other. The experiment is set up and we record 10 and zero. You can use the chi square formula, get a value and compare it to a critical value and is the number is greater that there is a statistical difference. If not than there is not statistical difference and the difference is said to be due to chance.

A hypothesis is an educated guess. A null hypothesis is one that states there is no statistically significant difference between the observed and expected values. So we can disprove our hypothesis if our data shows there to be a difference that is significant. Our working hypothesis is why we are doing the experiment in the first place. It states what one variable will do to the responding variable.

Hypotheses cannot be proven or correct. A null hypothesis allows us to state that we can reject that there is not a difference (which s a backward way of saying there is).

In summary, there are a variety of t test that allow us to find out how significant the differences are between two groups, linear regression allows us to find trends by drawing a straight line and chi square determines a goodness of fit between what is expect and what was observed.

We love you guys! Go out and be your best self today! And as always, Happy Nursing!