Chapter 4 Intro to Hypothesis Testing Lecture 4 Jaymie Ticknor Quantitative Methods 2317 Sect 001 31 January and 5 February 2014 Logic of Hypothesis Testing when conducting research generally use sample then want to see how results apply to a population Hypothesis testing is the procedure for deciding whether the outcome of a study using a sample supports a particular theory and applies to a population Theory more general set of principles that explain one or more facts relationships or events Hypothesis specific prediction that is tested in a study Hypothesis testing works by calculating the probability that the result we get is due to chance if this probability is very low we reject the idea that our experiment had no effect reject the Null Hypothesis opposite of actual hypothesis statement that in the population there is no effect or no difference between groups 1 2 If this probability is very high we fail to reject the idea that our experiment had no effect fail to reject the null hypothesis Hypothesis Testing Step 1 state the Null Hypothesis and Research Hypothesis statement that in the population there is an effect or there is a difference between groups 1 2 need to know about populations what populations are being compared Step 2 determine the characteristics of the comparison distribution hypothesis testing works by determining the probability the result we get could happen by chance another way to say this hypothesis testing works by determining the probability of getting our result if the null hypothesis is true must figure out some key information given about the comparison distribution Step 3 determine the cutoff sample score on comparison distribution where null hypothesis should be rejected researchers decide in advance how extreme a score must be for null hypothesis to be rejected use z scores and percentages Cultural backgrounds can have differences overall while molecules do not have much difference when they are under the same category so comparison would be stricter In psychology researchers generally use a cutoff score with probabilities of 5 and 1 conventional levels of significance false positive show up positive when actually negative Statistical significance when sample score is so extreme that researchers reject the null hypothesis Step 4 determine sample s score on the comparison distribution look at the actual study and figure out the z score Step 5 decide whether to reject the null hypothesis compare the sample z score with the cutoff z score if the sample z score is more extreme than the cutoff z score we reject the null hypothesis if the sample is not more extreme than the cutoff z score we fail to reject the null hypothesis When you fail to reject the null hypothesis results are not statistically significant inconclusive it could be that there was a difference but it was too small to be detected by the study Directional Hypothesis research hypothesis with a specific direction for the difference between groups one tailed test Nondirectional Hypothesis research hypothesis that does not predict a specific direction for the difference between groups two tailed test When using a two tailed test you have to divide the significance percentage between the two tails most researchers use two tailed tests more conservative procedure Hypothesis tests should be banned difference between statistical significance and street level importance Hypothesis tests should not banned uniqueness that can be used