Chapter 14 (Interpretation and Making Decisions About Data)

What are inferential statistics?

procedures that allow researchers to infer or generalize observations made with samples to the large pop. from which they were selected (about characteristics)

What is a null hypothesis (NHST)?

-A statement about a pop. parameter (such as the mean) that is assumed to be true
-null hypothesis significance testing
-a starting point
-we test whether the null value tends to be true

What is a test statistic?

-a mathematical formula that allows researchers to determine the likelihood of obtaining sample outcomes if the null hypothesis were true (can be used to make a decision regarding the null hypothesis)
-typically the p value > 0.05, we retain the null

When the null hypothesis is rejected, we reached _____.

significance

What are the two possibilities if we retain the null?

1. Null Result (Finding)- The value estimated in the null is true.
2. Type 2 Error- We say that the estimated value was true, but it actually is false in the population ("false negative")

What are the two possibilities if we reject the null?

1. Type 1 Error- "False Positive", controlled by researchers by stating their significance
2. Power- We decide that the null hypothesis was false, and it indeed was false (the probability that we will detect an effect if an effect actually exists in a pop

A researcher states that adult supervision during playtime sign. reduced aggressive behavior among children in her sample (p=.12). Is this conclusion appropriate at 0.5 level of significance?

No, because the p-value is greater than 0.05

What is the power or a decision-making process?

the probability of correctly rejecting a false null hypothesis

State the three assumptions or criteria that must be met to conduct a parametric test for a set of data?

-data in the pop. are normally distributed
-data are measured on an interval or ratio scale
-the variance of data in the pop. for each group is approx. the same or equal when many groups are observed

In a research study, one group of tasters was asked to taste and rate the sweetness of four different energy drinks on an interval scale measure. State the appropriate parametric test to analyze the data for this hypothetical study.

one-way within-subjects ANOVA

Which correlation coeff. is used to analyze data on an interval or ratio scale?

pearson correlation coeff.

Which two correlation coeff. are mathematically equiv. to pearson correlation coeff.?

-Spearman
-point-biserial
-phi correlation

What is a nonparametric test?

-significance tests that are used to test hypotheses about data that can have any type of distribution and to analyze data on a nominal or ordinal scale of measurement

What two conditions must be met to use a nonparametric tests to test hypotheses about parameters in a population?

-data in the pop. can have any type of distrib.
-data are measured on a nominal or ordinal scale of measurement

Know chart on 470.

...

What are the two types of nonparametric tests for nominal (categorical) data?

-chi- square goodness of fit test
-chi- square test for independence

What is a chi- square goodness of fit test used for?

-used to determine how well a set of observed frequencies fits with what was expected

What is a chi- square test for independence used for?

-to determine if frequencies at the levels of two categorical variables are indep. or related

State the nonparametric test alternatives to the following parametric tests:
-one sample t test
-two independent sample t test
-one way between subjects ANOVA

1. one-sample sign test
2. mann-whitney U test
3. kruskal-wallis H test

A health psychologist compares the number of students who are lean, healthy, overweight, and obese at a local school to expected proportions in each category. Should they use a chi-square goodness of fit test or a chi- square test for independence.

chi-sq. goodness-of-fit test

A researcher makes the decision to reject the null hypothesis using a chi-sq. test for independence. Does this decision mean that two factors are indep. or related?

the two factors are related

What is an effect?

a mean difference between what was observed in a sample and what was expected to be observed in a pop. (stated by a null hypoth.)

What is an effect size?

a stat. measure of the size or magnitude of an observed effect in a popp, which allows researchers to describe how far scores shifted in a pop., or the percent of variance in a dependent variable that can be explained by the levels of a factor

NHST indicates if a ______ exists in a pop. ______ indicates the size of an effect on a pop.

-effect
-effect size

What are the three effect size measures? LOOK AT TABLE ON PAGE 478

-Cohen's d
-proportions of variance (n2, R2)
-proportions of variance (Cramer's V)

What measure of proportion of variance is used to estimate effect size when data are analyzed using ANOVA?

eta-squared

Eta squared is mathematically equivalent to what other measure of proportion of variance?

the coefficient of determination

What test for significance uses Cramer's V to estimate effect size?

the chi-sq. test for independence

What is estimation? What are the two types?

-procedure in which a sample statistic is used to estimate the value of an unknown population parameter
-point estimation and interval estimation

What is point estimation?

A sample statistic (ex. a sample mean) used to estimate a population parameter (ex. a pop. mean).

What is an interval estimate? What is it called?

-the interval or range of possible values within which an unknown population parameter is likely to be contained

What are levels of confidence?

-The probability that an interval estimate will contain the value of an unknown population parameter is likely to be contained.
-has confidence limits (upper and lower boundaries)

If the null is inside the confidence interval, what is the decision?

-retain the null hypothesis (not significant)

What two rules do we use to identify the precision and certainty of a confidence interval?

1. Decreasing the level of confidence increases the precision of an estimate.
2. Increasing the level of confidence increases the certainty of an estimate.

A researcher compares health scores in a sample to the general health score known in the pop. and reports the following confidence interval: 95% CI (0.69-4.09). If the null hypoth. stated that there was no diff. in health scores between the sample and the

Retain the null hypothesis because the null hypothesis value of a mean difference equal to zero is contained within the stated confidence interval

What is the difference between the precision and certainty of a confidence interval?

The precision of an estimate is determined by the range of the confidence interval, whereas the certainty of an estimate is determined by the level of confidence

T of F: Selectively omitting data can lead to misleading interpretation of data.

True
-This is an ethical concern for "Full Disclosure of Data

CI are reported as:

-point estimate + and - an interval estimate