95 terms

A researcher is seeing if the level of satisfaction with college is different among students who participate in school activities compared to students who do not participate in school activities. Identify the dependent and independent variable in this study.

IV: participation in activities, DV: level of satisfaction

What are the two hypotheses that you state in step 1 of the hypothesis testing procedure called?

null hypothesis (h0), research hypothesis (h1)

Hypothesis testing is used with what general type of statistics?

inferential

In step 2 of the hypothesis testing procedure you determine the characteristics of the comparison distribution. Which population is the comparison distribution: Population 1 or Population 2?

population 2

To determine statistical significance you compare the z-score of your sample from Population 1 with what?

cutoff sample scores/critical values; z-scores of % looked up in tail

Indicate whether you would use a one-tailed or two-tailed test for the following hypothesis: Netflix had fewer subscribers after they raised their prices.

1-tailed

Why do researchers tend to use two-tailed tests rather than one-tailed tests?

one-tailed test can make you miss a statistical significance if it on the other end of spectrum

State the following research hypothesis in symbols: Nebraska Wesleyan students have higher ACT scores than UNL students.

u1 > u2

In which of the 5-steps of hypothesis testing do you indicate whether you test is one- or two-tailed?

step 3

What type of distribution has more variance?

distribution of individuals

A distribution of means is most likely to have what shape?

normal

When your sample consists of a group of people (population 1) why can you not compare that to a distribution of individuals (population2)?

because you must compare like with like and not groups to individuals

Which decision error is generally considered by researchers to be the more serious error

type I

Why is there always the possibility when using statistics that our conclusions will be wrong?

there is always some chance of error because we are never 100% sure

The significance level for a particular statistical test is .05. Which type of decision error will also have a 5% likelihood of occurring

Type I

True or False? You can make both a Type I and Type II error in the same study?

false

A psychologist who states the hypothesis that there will be no difference between a new approach to therapy and the standard approach is stating a

null hypothesis

When a psychologist rejects the null hypothesis at the .05 level, the result of a study indicate that

there is less than a 5% chance of getting such an extreme result by chance if the null hypothesis is true

A result is considered statistically significant when

the sample score is so extreme that the null hypothesis is rejected

The decision to reject the null hypothesis using the Z test is made

by comparing the Z score needed to reject the null hypothesis to the actual Z score of the sample

A psychologist who states the hypothesis that a new training program will be more effective than the old training program is stating a

research hypothesis

How would a psychologist test the hypothesis that a new stress reduction program really works?

try to reject the hypothesis that it does not work

The distribution that represents the situation in which the null hypothesis is true is

the comparison distribution

If the cutoff Z score on the comparison distribution is +/- 2.31, a psychologist can reject the null hypothesis if the sample Z score on this distribution is

2.83

When the results of a study are not extreme enough to reject the null hypothesis, a psychologist can conclude with reasonable confidence that

the results are inconclusive

If a health psychologist wants to know if regular exercise improves peoples' health, the type of hypothesis the psychologist would use is

one-tailed because the study is only interest in whether the exercise increases health

The correct argument for using a one-tailed test when there is a clear basis for predicting a result in a given direction is that

it is less conservative in that one-tailed tests make rejecting the null hypothesis easier

A major misuse of significance tests is the tendency to

decide that if a result is opposite to the prediction, the researcher can still do a two-tailed test later

A major misuse of significance tests is the tendency to

decide that if a result is not significant, the null hypothesis is shown to be true

Which of the following statements is the most accurate description of hypothesis testing?

it is a central theme in the statistical analysis of virtually all psychology research

When your sample is a group of people the correct comparison distribution to use is a distribution of means because

comparing the mean of a sample to a distribution of individuals is a mismatch

In principal, a distribution of means can be formed by

randomly selecting samples from a population, finding the mean of each sample, and then graphing the means

The mean of a distribution of means is

the same as the original population mean

The variance of a distribution of means is smaller than the original population variance because

extreme scores are less likely to affect a distribution of means

In general, the shape of a distribution of means tends to be

unimodal & symmetrical

A study involving 10 participants has a known population mean of u=100 and variance of 25. Using the distribution of means, what characteristics of the comparison distribution would you report in Step 2?

mean = 100, variance = 2.5

As the number of people in a sample gets larger, the distribution of means

becomes a better approximation of the normal curve

Under what conditions is it reasonable to assume that distribution of means will follow a normal curve

30 or more individuals involved or distribution of population means is normal

One important advantage of using effect sizes is that

they are standardized scores that make comparisons of different studies easier

The alpha level is

the probability of a type I error

If a study conducted at the .05 significance level has 80% power

alpha = 5%, beta = 20%

In actual practice, the usual reason for determining power before conducting a study is

to determine the number of participants needed to have a reasonable level of power

Type II errors concern psychologists because

they could mean that good theories or useful practical procedures are not used

Setting the significance level cutoff .10 instead of the more usual .05 increases the likelihood of

Type I error

Beta is the probability that

if the research hypothesis is actually true, the experiment would still fail to support it

The statistical method currently used to combine the results of multiple studies is

meta-analysis

A type I error is the result of

incorrectly rejecting the null hypothesis

Power is the probability that

if the research hypothesis is true, the experiment will support it

Failing to reject the null hypothesis when the research hypothesis is true is

type II error

Effect size is

the degree to which an experimental manipulation separates two populations

If the research hypothesis is true, but the study has a low level of power,

the probability that the study will have a significant result is low

Cohen proposed effect size conventions based on the effects observed in psychology research in general because

previously it was difficult to determine how large an effect size was

The conventional levels of significance of 5% and 1%

are a compromise between the risk of making Type I and Type II errors

hypothesis testing

procedure for deciding whether the outcome of a study supports a particular theory

hypothesis

prediction, often based on informational observation, previous research, or theory that s tested in a research study

theory

set of principles that attempt to explain one or more facts, relationships, or events; predictions are derived from theories

research hypothesis

statement in hypothesis testing about the predicted relation between populations

Null hypothesis

statement about a relation between populations that is the opposite of the research hypothesis

comparison distribution

distribution used in hypothesis testing; represents the population situation if the null hypothesis is true

cutoff sample score

in hypothesis testing, the point on the comparison distribution at which, if reached or exceeded by the sample score you reject the null hypothesis

conventional levels of significance

p<.05, p<.01; levels of significance widely used in psychology

statistically significant

conclusion taht the results of the study would be unlikely if in fact the sample population is no different from the general population

directional hypothesis

research hypothesis predicting a particular direction of difference between populations

one-tailed test

hypothesis-testing procedure for a directional hypothesis; region of comparison distribution in which the null hypothesis would be rejected is all on one side of the distribution

Nondirectional hypothesis

research hypothesis that doesn't predict a particular direction of different between populations

Two-tailed test

hypothesis-testing procedure for a nondirectional hypothesis; region of the comparison distribution in which the null hypothesis would be rejected is divided between the two sides of the distribution

distribution of means

distribution of means of samples of a given size from a population; comparison distribution when testing hypotheses involving a single sample of more than one individual

mean of a distribution of means

the means of a distribution of means of samples of a given size from a population; same as u

variance of a distribution of means

variance of the population divided by the number of scores in each sample

standard deviation of a distribution of means

square root of the variance of a distribution of means; aka SEM or SE

decision errors

incorrect conclusion in hypothesis testing in relation to the real situation

Type I error

rejecting the null hypothesis when in fact it is true

Alpha

probability of making a Type I error; same as significance level

Type II error

failing to reject the null hypothesis when in fact it is false

Beta

probability of making a Type II error

Effect Size

in studies involving means of one or two groups, measure of difference between populations

Effect size conventions

standard rules about what to consider a small, medium, & large effect size, based on what is typical in psych research

Meta-analysis

statistical method for combining effect sizes from different studies

Statistical power

probability that a study will give a significant result if the research hypothesis is true

power tables

table for a hypothesis-testing procedure showing the statistical power of studies with various effect sizes & sample sizes

Be able to explain the relationship between effect size and statistical significance in interpreting research results

they are unrelated because effect size is how big the difference is between our means (practical significance) and statistical significance is if the is a difference between our groups (hypothesis testing; decision errors); power is related to both effect size and statistical significance

Independent variable

what is being manipulated, predictor, cause

Dependent Variable

what is being measured, what is being predicted, effect

Steps of Hypothesis Testing

1. Identify populations & hypotheses 2. Determine population mean, population standard deviation 3. Determine cutoff sample score on comparison distribution at which the null hypothesis should be rejected 4. Determine your sample's score 5. Decide whether to reject the null hypothesis

p<.05 are what cutoff scores for 1-tailed & 2-tailed

1.64 & 1.96

<0.2 effect size

no effect

0.2-0.49

small effect

0.5-0.79

medium effect

>/= 0.8

large effect

What determines power?

sample size (larger sample increases power), effect size (larger effect size increases power), significance level (power is higher @.5), one vs. two-tailed test (varies power)

Use of power table

determining # of participants before study, interpreting results of a study

low power with non significant results means

truly inconclusive

high power with non significant results means

unlikely that the research hypothesis is true or small effect

lower power with significant results means

likely practical significance

high power with significant results means

may not be practically significant