Psychology 395 Research Methods In Psychology Exam Four Statistical Inference

Front 1

Central Limit Therum

Back 1

If something is sampled enough times, data will fall within +or- 2 measures of the sample mean all but 4% of the timte.

Front 2

Type I Error (False Positive)

Back 2

Incorrectly identifying a difference as significant when it was not. Probability of making this mistake is Alpha(.05). The data has no significant difference.

Front 3

Correct Positive Decision

Back 3

Correctly assessing that there is significant difference. Probability of this occuring is one minus beta (P= 1-B)

Front 4

Correct Negative Decision

Back 4

Correctly assessing that there is no significant difference. Probability of this occuring is one minus alpha (P= 1-A)

Front 5

Type II Error (False Negative)

Back 5

Incorrectly failing to identify a significant difference that actually exists. (not realizing there is a significant different) Probability of this occuring is Beta (P = B)

Front 6

Statistical Power

Back 6

The probability of correctly detecting a difference or relationship that actually exists. Probability of this occuring is one minus Beta (P= 1-B)

Front 7

Statistical Power is a Function of:

Back 7

1)Sample Size
2)Variance in the population (on a measured variable)--less population variance increases the statistical power.
3)Effect Size-Difference between the population mean and the sample mean (strength of manipulation)
4)Alpha Level-increases power, but at the cost of lower number of correct negatives, increases false positive.
5)Reliability of Measurement

Front 8

Correct Positive Decisions

Back 8

Research advances with the discovery of correct positive decisions.

Front 9

Multivariate Design

Back 9

A study with two or more measured variables.

Front 10

Multivariate Statistical Test

Back 10

For a study with two or more variables which:
1)Test of significance (of differences or relationship)
2)An index to check ovarall probability of differences considering all at once
3)For use before reporting results for univariate tests.

Front 11

Null Hypothesis (H o)

Back 11

Accepting hypothesis means that z < the absolute value of 1.96
Rejecting means z > the absolute value of 1.96

Front 12

Analysis of Variance (ANOVA)

Back 12

Data-analysis using a kind of inferential statistic, the F-test, to detect differences among >(or equal to) 2 comparison groups defined by one of more factors, and if two or more, their interactions. Ex. in analyzing results of an experiment on effects of three audience sizes on bystanders' reponse-time, using an F-Test to compare the averages for the three experimental groups.

Front 13

Inferential Statistics

Back 13

Numerical indexes computed to indicate the likelihood that observed differences among groups, or relationships between variables, did not occur by chance
Ex. Z-test, Pearson r. (inferential statistics test the likelhood that one or more sames came from the same population, the basis for statistical significance.

Front 14

Standard Error of the Mean

Back 14

Standard deviation of the distribution of the sample mean for an infinite number of samples of size (n) drawn from a population with a known mean for individual scores of (U) and known standard deviation of indiviual scores of (o).
Ex. In a pop of college students whose GRE Verbal scores have a standard deviation of 100 for samples of 25 students, and the standard error of the mean GRE Verbal score is (100/5=20).