Textbook Notes (368,434)
Canada (161,878)
Psychology (2,971)
PSY201H1 (45)

Chapter Ten PSY201

11 Pages
Unlock Document

Kristie Dukewich

Chapter 10: Introduction to Hypothesis Testing Using the Sign Test inferential statistics has two main purposes + 1. hypothesis testing + 2. parameter estimation Logic of Hypothesis Testing Marijuana and the Treatment of AIDS Patients data could be analyzed with several different statistical inference tests such as sign test + easy to understand + all the major concepts concerning hypothesis testing illustrated simply and clearly + ignores the magnitude of difference of scores and considers only their direction/sign + omits a lot of information which makes the test rather insensitive + if consider only the signs of the difference scores, then experiment produced 9/10 pluses amount of food eaten in experimental condition was greater after taking the THC pill in all but one of patients suppose that marijuana has absolutely no effect on appetite + is it possible to have obtained 9/10 pluses? Yes + if marijuana has no effect on appetite, then each subject would've received two conditions that were identical except for chance factors + perhaps when subject 1 was run in THC conditions, he slept better night before and his appetite was higher than when run in the control condition before any pills were taken expect him to eat more food in THC condition even if THC has no effect on appetite + or perhaps subject 2 had a cold when run in placebo condition which blunted her appetite relative to when run in experimental condition expect more food to be eaten in experimental condition even if THC has no effect could go on giving examples for other examples + point: these explanations of greater amount eaten in THC condition are chance factors + they're different factors, independent of one another, and could just as easily occurred on either of the two test days + seems unlikely to get 9/10 pluses simply as a result of chance factors + How unlikely is it? if we know chance alone is responsible, we'd get 9/10 pluses only 1 time in 1 billion rare occurrence; reject chance and the explanation that marijuana has no effect on appetite. Conclude by accepting the hypothesis that marijuana affects appetite because it's the only other explanation - since sample was a random one, can assume it was representative of AIDS patients being treated at hospital generalize the results to that population probability of getting 9/10 pluses due to chance alone is really 1 in 3, not 1 in 1 billion. + decision to reject chance as a cause of data is not clear + instead, need a rule for determining when the obtained probability is small enough to reject chance as an underlying cause Repeated Measures Design repeated measures, replicated measures, or correlated groups design + essential features are that there are paired scores in the condition and the differences between paired scores and analyzed in marijuana experiment, used the same subjects in each conditions subjects served as their own controls + their scores were paired, and the differences between these pairs were analyzed + instead, in same subjects, could've used identical twins or subjects who were matched in some other way two conditions + experimental and control condition + two conditions are kept as identical as possible except for values of the independent variable which are intentionally made different + in our example, marijuana is the independent variable Alternative Hypothesis (H1) there are two hypothesis that compete for explaining the results + alternative hypothesis and null hypothesis alternative hypothesis is the one that claims the difference in results between conditions is due to the independent variable + hypothesis claims marijuana affects appetite + can be directional or non-directional + hypothesis marijuana affects appetite is non-directional because it doesn't specify the direction of the effect + if hypothesis specifies the direction of the effect, it's a directional hypothesis marijuana increases appetite = directional alternative hypothesis Null Hypothesis (H0) logical counterpart of alternative hypothesis + if false, then alternative hypothesis must be true + these two hypothesis must be mutually exclusive and exhaustive if alternative hypothesis is non-directional, it specifies that the independent variable has an effect on the dependent variable + for this non-directional alternative hypothesis, the null hypothesis asserts that the indep
More Less

Related notes for PSY201H1

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.