

Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Community
Ask the community for help and clear up your study doubts
Discover the best universities in your country according to Docsity users
Free resources
Download our free guides on studying techniques, anxiety management strategies, and thesis advice from Docsity tutors
PSYCH 350 Exam 2 Essays correctly answered 2022.
Typology: Exams
1 / 2
This page cannot be seen from the preview
Don't miss anything!
Describe the process of NHST, tell the (five) possible outcomes and tell the likely reasons for each. (Be sure to tell what this acronym means.) HIGH PROBABILITY - correct answer The process of NSHT (null hypothesis significance testing) is when we get a sample. From that sample we collect data, which is analyzed and we get a test statistic out of it. From that test statistic we get a p-value, which is used to decide whether or not there is a relationship between the population. Afterwards, you replicate the study which is followed by programmatic research. Two outcomes are good, three are bad. The good is a correct retention of the null hypothesis and we could have a correct rejection because p is less than .05, meaning there is a relationship between the two variables. The bad are type I error, false alarm (when we think we found something but actually didn't), type II, miss, (when there is a relationship but researchers missed it) and type III, misspecification, (when we find a correlation or effect but it is opposite to that of our hypothesis). All three are caused by bad samples, bad measurements, confounds, and last is chance. Type II also includes small samples. Tell when to use each type of ANOVA, the possible research hypotheses for this statistical model, and when ANOVA can be used to test each type of Research Hypothesis (attributive, associative and causal). MEDIUM - correct answer Between groups anova are used with between groups designs, within group anova uses within group designs. The possible research hypothesis are always about mean differences. Either group one does poorer than group two or they are equal, or group one does better than group two. Research hypothesis can be null hypothesis. Attributive, never! Associative, always! Anytime you are doing an ANOVA, you are at least evaluating an associative relationship. Causal, depends on the design, if it is a true experiment without confounds and properly ran, the results are causally interpret-able. Tell when to use a Pearson's correlation, the possible research hypotheses for this statistical model, and when correlation can be used to test each type of Research Hypothesis (attributive, associative and causal). HIGH PROBABILITY - correct answer When both variables are quantitative and RH is about their linear relationship. Regarding the possible research hypotheses for this statistical mode: The correlation is positive, zero or negative. RH can be the null. Attributive, never!, Associative, always and most commonly tested by correlation! Causal, depends if a true experiment. Tell when to use Pearson's X², the possible research hypotheses for this statistical model, and when X² can be used to test each type of Research Hypothesis (attributive, associative and causal). - LOW PROB. - correct answer When we have two qualitative variable and we want to know if they are related. Regarding the possible research hypothesis for this statistical model: They come in specific pattern to the data, or no pattern. RH can be null. Attributive, never(attributive is univari and chi-square is
bivariate)! Associative, Always(if two variable are related)! Causal, depends if it is a true experiment. Compare and contrasts the "interesting pairs" of the four bivariate data analysis models we are working with. - LOW PROB - correct answer Correlation(two quantitative) and chi-square(two qualitative) both look at bivariate relationship. Between groups F(two different groups have different means) and within group F(same group at different times have different means) which look for means differences. Between group F and chi- square, both are between groups design comparison, anova is means differences and chi-square pattern of responses. Correlation and the within groups F both use two quantitative variables the correlation ask for linear relationship the anova ask about mean difference between the variables, You can't always run both an r and within group F on the same two quantitative variables. Respond to and describe the statement, "Rejecting the null hypothesis guarantees support for the research hypothesis." - HIGH PROB - correct answer No!!!! First reasoning is the RH might be the null hypothesis. An anova, correlation, and chi-square are all allowed to hypothesize that there is no relationship between the variables. If the research hypothesis is the null, rejecting the null isn't supporting it. Second, when you reject the null but find a data pattern different from the RH. Example: you hypothesize a positive correlation but you find a significant correlation. The fact that it's significant means that you are gonna reject the null but a negative correlation doesn't support the hypothesis of a positive correlation Describe effect size estimates, tell how they are related to significance tests, and the information they provide that is not provided by significance tests. - MEDIUM - correct answer Effect sizes talk about the strength of the relationship between the variables. NHST tells about the likelihood of a relationship between the variables, if there is, it is reasonable to ask how strong which you get from the effect size estimate. the effect size estimate is essentially a significant test with the sample size removed. It is to make sure if we find the same correlation that we notice it. We don't want the effect size influenced by the sample size. We want the null hypothesis significance test to be influenced by the sample size because bigger studies are more likely to be accurate What is meant by "statistical power" and what is the advantage if our research has lots of it? Describe how power analyses are conducted and how they can inform our statistical decisions. - HIGH PROB. - correct answer Statistical power aka sensitivity is the ability to reject the null when the null is wrong in the population. It is the ability to find an effect when an effect is present in the population. The advantage if there is lots of power, it means few type II errors. If we have lot's of power, we don't miss effects. Power analyses are conducted by a priori which is conducted before the study to figure out proper sample size. Also, Post Hoc, which is conducted after the study and when we have retained the null. We do this to estimate the probability that we just committed a type 2 error with our retained null.