CH13_Why We Need Statistics

Question 1

What is the term for making a statement about the population and all its samples based on what we see in the samples we have?

Answer 1

Statistical inference.

Question 2

A statement that the data came from different populations; the research hypothesis, which cannot be tested directly

Answer 2

Alternative hypothesis

Question 3

The portion in the tail(s) of the distribution of a test statistic extreme enough to satisfy the researcher’s criterion for rejecting the null hypothesis—for instance, the most extreme 5% of a distribution where p

Answer 3

Critical region

Question 4

The standard procedures used to summarize and describe data quickly and clearly; summary statistics reported for an experiment, including mean, range, and standard deviation

Answer 4

Descriptive statistics

Question 5

A statement that predicts the exact pattern of results that will be observed, such as which treatment group will perform best

Answer 5

Directional hypothesis

Question 6

Variation in subjects’ scores produced by uncontrolled extraneous variables in the experimental procedure, experimental bias, or other influences on subjects not related to effects of the IV?

Answer 6

Experimental error

Question 7

Statistics that can be used as indicators of what is going on in a population; also known as test statistics

Answer 7

Inferential statistics

Question 8

An arithmetical average computed by dividing the sum of a group of scores by the total number of scores; a measure of central tendency

Answer 8

Mean

Question 9

The score that divides a distribution in half, so that half the scores in the distribution fall above it, and half below; a measure of central tendency

Answer 9

Median

Question 10

The most frequently occurring score in a distribution; a measure of central tendency

Answer 10

Mode

Question 11

A statement that predicts a difference between treatment groups without predicting the pattern of results

Answer 11

Non-directional hypothesis

Question 12

The distribution of data in a symmetrical, bell-shaped curve

Answer 12

Normal curve

Question 13

A statement that the performance of treatment groups is so similar that the groups must belong to the same population; a way of saying that the experimental manipulation had no important effect

Answer 13

Null hypothesis (H0)

Question 14

A statistical procedure used when a directional prediction has been made; the critical region of the distribution of the test statistic is measured in just one tail of the distribution

Answer 14

One-tailed test

Question 15

The difference between the largest and smallest scores in a set of data; a rough indication of the amount of variability in the data

Answer 15

Range

Question 16

Data recorded as an experiment is run; the responses of individual subjects

Answer 16

Raw data

Question 17

The statistical criterion for deciding whether to reject the null hypothesis or not, typically p

Answer 17

Significance level

Question 18

The square root of the variance; measures the average deviation of scores about the mean, thus reflecting the amount of variability in the data

Answer 18

Standard deviation

Question 19

Meeting the set criterion for significance; the data do not support the null hypothesis, confirming a difference between the groups that occurred as a result of the experiment

Answer 19

Statistical significance

Question 20

Quantitative measurements of samples; quantitative data

Answer 20

Statistics

Question 21

Descriptive statistics computed from the raw data of an experiment, including the measures of central tendency and variability

Answer 21

Summary data

Question 22

Statistics that can be used as indicators of what is going on in a population and can be used to evaluate results; also called inferential statistics

Answer 22

Test statistics

Question 23

A statistical procedure used when a nondirectional prediction has been made; the critical region of the distribution of the test statistic is divided over both tails of the distribution

Answer 23

Two-tailed test

Question 24

An error made by rejecting the null hypothesis even though it is really true; stating that an effect exists when it really does not

Answer 24

Type 1 error

Question 25

An error made by failing to reject the null hypothesis even though it is really false; failing to detect a treatment effect

Answer 25

Type 2 error

Question 26

Fluctuation in data; can be defined numerically as the range, variance, or standard deviation

Answer 26

Variability

Question 27

The average squared deviation of scores from their mean; a more precise measure of variability than the range

Answer 27

Variance

Question 28

T/F: When we do a statistical test, we are asking whether our pattern of results is significantly different from what we would expect to see given the usual variability among different people in the population.

Answer 28

T.

In psychology, we accept the outcome of statistical tests to establish whether an independent variable had an effect in a particular experiment.

Question 29

T/F: We assume data from different treatment groups came from the same population, and any differences between them amount to nothing more than the ordinary variability in scores we would expect in any population.

Answer 29

T.

Question 30

If we reject the null hypothesis, we are confirming a change between the groups that occurred as a result of the experiment: we say that our results are ______.

Answer 30

Statistically significant.

Question 31

What is a way to test the alternative hypothesis?

Answer 31

There is no way to directly test the H1 (or the research hypothesis). Therefore, we can never really PROVE that our research hypothesis is correct. The best we can do is show that it is unlikely that the pattern occurred from chance variation within the population we sampled. We can only show H0 is PROBABLY wrong.

Question 32

How can we phrase indirectly testing the alternative hypothesis?

Answer 32

We attempt to show that the null hypothesis is probably false.

Question 33

T/F: The less variability, the harder it will be to reject the null hypothesis.

Answer 33

F.

The more variability, the harder it will be to reject the null hypothesis.

Question 34

After you compare two means, and let’s say the higher mean would imply you can reject the null hypothesis, should you already reject the null hypothesis?

Answer 34

No. You have to consider variability in the data, and evaluate these with statistical tests.

Question 35

T/F: Many of the statistical tests used by psychologists include the assumption that the population you have sampled is normally distributed on the dependent variable.

Answer 35

T.

Question 36

T/F: In psychology, by convention, we generally reject the null hypothesis if the probability of obtaining this pattern of data by chance alone is less than 5%

Answer 36

T.

We then say the significance level is p

Question 37

T/F: The significance level must be decided before running an experiment.

Answer 37

T.

It is not legitimate to collect all the data and then pick the significance level depending on how the results turned out. It would yield significant results, but only because we stacked the deck in our favor.

Question 38

T/F: If our results are statistically significant, there is no way the null hypothesis could be true.

Answer 38

F.

Even in those cases, there is always some probability that the null hypothesis could still be true.

Question 39

T/F: The odds of making a Type 1 error are equal to the value we choose as the significance level for rejecting the null hypothesis.

Answer 39

T.

For example, if we are using a .05 significance level, the probability of a Type 1 error is .05. If our significance level is .05. then 5 times out of 100 (or 5% of the time), we will reject the null hypothesis when we should not.

Question 40

What is the risk of minimizing the odds of a Type 1 error by choosing a more extreme significance level?

Answer 40

It becomes more possible to make a Type 2 error. In other words, you can fail to reject the null hypothesis even if it is really false. You can miss a treatment effect that was really present.

Question 41

T/F: The probability of making a Type 2 error is affected by the amount of overlap between the populations being sampled.

Answer 41

T.

In other words, if responses from the different treatment groups are very similar, you are prone to making a Type 2 error.

Question 42

The probability of making a Type 2 error is represent by the Greek letter _____

Answer 42

Beta.

Question 43

T/F: The closer a p-value is to 0, the more confidence we have to reject the null.

Answer 43

T.

Question 44

T/F: The concept of the p-value at .05 is that if there is no difference between two treatments, and if we did the exact same experiment a number of times, then only 5% of those experiments would result in the wrong decision.

Answer 44

T.

Question 45

T/F: You can have a small p-value regardless of the size of difference between Drug A and Drug B.

Answer 45

T.

A small p-value does not imply that the effect size, or difference between Drug A and Drug B is large.

Question 46

What does ‘statistically significant’ mean?

Answer 46

It means there was enough evidence to reject a null hypothesis.

Question 47

What two factors affect the odds of finding significance?

Answer 47

The amount of variability in the data and whether we have a directional or a nondirectional hypothesis.

Question 48

T/F: The wider a distribution is, the larger the variability.

Answer 48

T.

Question 49

T/F: If the population has high variability, finding a statistically significant effect requires very large differences between the mean scores from different treatment groups in the experiment.

Answer 49

T.
Therefore, any unnecessary sources of variation in an experiment will not only reduce the chances of rejecting the null hypothesis, they will also increase the chances of a Type 2 error. It will be difficult to get significant results if extraneous variables are not carefully controlled.