PSY Exam 2

one-sample t-test

- compare sample mean to population mean OR meaningful score (comparing to chance, 50%) - don't know population SD (O) only know sample SD (s)

2. The one-sample z test is a hypothesis test used to test hypotheses ______. A. concerning a single population with a known variance B. concerning at least one population C. concerning the variance in a population D. all of these

A. concerning a single population with a known variance

8. A researcher records the number of distracters (such as noises) that preschool-aged children ignore while watching a popular Sunday morning cartoon show. Assuming that the population variance is unknown, what type of t test is appropriate for this study? A. one-sample t test B. two-independent sample t test C. There is not enough information to answer this question.

A. one-sample t test

13. A type of related-samples design in which participants are observed more than one time is called a ______. A. repeated-measures design B. matched-pairs design

A. repeated-measures design

A

Assumptions: check your assumptions NHST makes several assumptions that must be met before you can do the test t-test assumptions: - random sampling - independence of observations -normality (same as for z-test

6. A researcher conducts two studies. Each study was a one-sample z test. Both studies placed the rejection region in both tails and measured the same sample mean. The beat level in Study 1 was larger than the beta level used in Study 2. Which study is associated with greater power to detect an effect? A. Study 1 B. Study 2 C. They are associated with the same level of power.

B. Study 2

3. A researcher obtains z = -6.45. What is the decision for a one-tailed test, upper-tail critical, at a .05 level of significance? A. to reject the null hypothesis B. to retain the null hypothesis C. It depends on the sample size. D. There is not enough information to make a decision.

B. to retain the null hypothesis

1. If a researcher obtains a null finding, then what is the decision? A. They correctly rejected the null hypothesis. B. They incorrectly rejected the null hypothesis. C. They correctly retained the null hypothesis. D. They failed to make a decision.

C. They correctly retained the null hypothesis.

5. Based on the effect size conventions, d = 0.90 is a ______. A. small effect size B. medium effect size C. large effect size

C. large effect size

C

Calculate: calculate the test statistic

4. 3A measure of the size of an effect in a population is called ______. A. significance B. probability C. power D. effect size

D. effect size

D

Decision rule: specify significance level and critical values - how different is different? - p-values are the probability that the difference we see is just due to random chance and not an actual difference - we want small p-values - the significance level (a) is the maximum p-value we will accept in order to conclude that something is significantly different from chance - usually set to a=.05

10.True or Fale: A t test is associated with N degrees of freedom.

False

H

Hypothesis: write your null/alternative hypothesis (required both) hypothesis must be: - written numerically - exhaustive (every possible option is included) - mutually exclusive (can not overlap) - written about the population EX: H0: u > 2.1 -> Cara's class reads the same amount books as the rest of the school H1: u < 2.1 -> Cara's class reads less books than the rest of the school

I

Interpretation: make a decision and write up the results - two things you can do with H0 -reject or fail to reject - reminder: H0 = no difference - reject -> there is a difference - fail to reject -> not enough evidence to say there's a difference NOTE: failing to reject is NOT the same as saying the two values are the same, you do not "accept" the H0

T.A.H.D.C.I

TA Henry's Dog Crapped Inside

T-

Test: pick which statistical test you are going to do (z,t, ANOVA)

independence of observations

Two observations are independent if the occurrence of one observation provides no information about the occurrence of the other observation.

Calculating CI's

X with line over: sample mean +/-: plus and minus tcrit: critical value of t sx: estimated standard error of X

Confidence Interval (CI)

a range of scores that we are _____ % confident contains the actual pop. mean - typically use 90%, 95%, or 99% - CI: the proportion/percent of possible CI that ontain the true value of a population parameter if the study were exactly repeated an infinite number of times - convey how precise our point estimates are - if we were to conduct our study 100 times (get a sample mean) we expect in the long run 95 times to have the pop. mean difference within our interval

For bigger df

always round down (bigger tcrit -> more conservative test

Region rejection

any t-value more extreme than the tcrit is considered to be in the rejection region - reject the H0

paired samples t-test

compare two RELATED sample means to see if there's a difference (don't know pop. M/SD for either group -> only know sample M/SD

independent samples t-test

compare two sample means to see if there's a difference - don't know pop. M/SD for either group -> only know sample M/SD (called between-groups design)

between subjects design

compare two sets of data taken from different sets of people

within subjects design

compare two sets of data taken from the same set of people (also called repeated measures design)

Interpretation example:

format: "t(df)=t-value, p-value" We reject the null hypothesis. The student's in Cara's class M(=1.9) read significantly less books than the rest of the student's in the school (M=2.1; t(35)=171, p < .05).

The T-Distribution continued

has larger tails than the standard normal distribution (z) - this makes sense bc knowing less about the population makes us be more conservative - SMALL df means large tails - LARGE df means small tails As df becomes large the t-dist approaches the z distribution

matched-pairs design

matching participants based on academic ability

CI's and NHST

the CI can tell us whether or not to reject the H0 - if the CI does NOT contain the pop. mean (reject the H0) - If the CI contains the pop. mean (fail to reject H0) - the CI will always give you the same result as the NHST

homogeneity of variance

the variances of the two groups are roughly the same

natural pairs

twins

normality

used to determine if a data set is well-modeled by a normal distribution and to compute how likely it is for a random variable underlying the data set to be normally distributed.