Mass Media Research: An Introduction, 8th Edition Instructor’s Manual (стр. 8 )

d)  May elect to collect more data before a decision

7.  It is common practice in mass media research studies to set the probability level used in hypothesis testing to_____.

a)  .25

b)  .95

c)  .10

d)  None of the above

8.  In a two-tail test, the region of rejection:

a)  Is located in the right tail of the sampling distribution curve

b)  Is located in the left tail of the sampling distribution curve

c)  Is located in both the left and right tails

d)  Cannot be precisely determined

9.  The power of a statistical test refers to:

a)  How large a sample size is needed

b)  The level of measurement used

c)  The probability that a statistical test will result in a Type I error

d)  The probability that a statistical test of a null hypothesis will result in the conclusion that the phenomenon under study actually exists

True/False

1.  In hypothesis testing, the researcher rejects or accepts the alternative hypothesis. (F)

2.  A one-tail test is more stringent than the two-tail test. (T)

3.  There is no easy answer to the problem of balancing Type I error and Type II error, but power analysis to help researchers deal with this issue. (T)

4.  Many researchers suggest a desirable power value is.8 when working at the.05 level of significance. (T)

Fill in the Blank

1.  (Exploratory) research is intended to search for data indications rather than to attempt to find causality.

2.  The (null hypothesis) asserts that the statistical differences or relationship being analyzed are due to chance or random error.

3.  In a theoretical sampling distribution, the proportion of the area in which the null hypotheses is rejected is called the (region of rejection).

4.  (Type I error) is the rejection of a null hypothesis that should be accepted.

5.  Type II error often called (beta error).

Short Answer

1.  Explain Type I error and Type II error by using an example.

2.  A medical research group tried to produce new medicine for curing AIDS. One researcher used a probability level of.05 when he tested a new medicine and another researcher in that group used a probability of level of.001 when testing a second medicine. Assume both medicines showed significant results. If you were a research consultant, which medicine would you recommend to the customer? Why?

Chapter 12 - Basic Statistical Procedures

Overview

This chapter discusses basic statistical procedures. Because statistics are necessary to understand the scientific method of knowing because they allow researchers to make inferences about the population from which the sample has been taken. Specially, this chapter describes the basic inferential statistical methods (chi-square goodness of fit, contingency table analysis, t-test, analysis of variance, correlation, and regression) used in mass media research and suggests ways in which these methods may help answer research questions.

Statistical methods are commonly divided into two broad categories: nonparametric and parametric. In the past, both types of statistics had some distinctions based upon levels of measurement. For the most part, the distinctions have vanished. Both nonparametric statistics and parametric statistics can be used successfully with all types of data and that both are appropriate for generalizing results to the population when used with a random sample.

For nonparametric statistics, mass media researchers often compare the observed frequencies of a phenomenon with the frequencies that might be expected or hypothesized. This is the Chi-square goodness of fit test. In addition, this chapter also discusses cross-tabulation, or contingency table analysis.

Parametric statistics defined by two characteristics: 1) these methods usually use interval and ratio data and 2) these methods assume that data are normally distributed. This chapter introduces that t-test as the most elementary method for comparing two groups’ mean scores.

For comparing more than two groups’ mean scores, the chapter introduces analysis of variance (ANOVA). This statistic can be used to simultaneously investigate several independent variables, (also called factors). A one-way ANOVA investigates one independent variable; a two-way ANOVA investigates two independent variable, and so on. In two-way ANOVA, researchers may find main effects and interaction effects in the relationships among the variables.

To understand the relationships between two variables in a study, researchers can use correlational statistics: numerical expressions of the degree to which two variables change in relation to each other. If researchers want to control confounding variables in the relationships between two variables, they may use partial correlation, eliminating the effect of spurious variables.

Regression is introduced in the chapter. Based on correlation between variables, this statistic is used to determine the degree to which one variable changes with a given change in another variable. If researchers deal with the relationship between two or more independent variable and a single dependent variable, this statistic is called multiple regression.

Exercises

Borrow basic statistics textbooks from the library and check the mathematical logic of each nonparametric and parametric statistic. Fisher introduced the idea of the likelihood that an event will occur. How likely is it that a pair of sixes will come up when you roll two dice?

Practical Problems

Suppose the researcher finds a chi-square value of 11.71, with degrees of freedom of 4, and has established an alpha level of.01. What do you expect that the researcher will do concerning the null hypothesis? Explain why. The following are data of Internet use per day by gender. Using t-test, determine the statistical significance with probability.05 between the two groups. Are these groups statistically different or not? Why?

Gender Male Female

Mean 2 hours 6 hours

Participants 10 10

Standard error 57 51

Answers:

The Chi-square table shows a value of 13.277 at this level (df = 4 and probability = .01). Since 11.71 is smaller than 13.277, the difference is not significant. t = X1 – X2/Sm t = 2 – 6 / 1.09 = 3.67 df = 18, t =3.67

If the problem is tested at the.05 level of significance, a t value of 2.101 is required for the results to be considered statistically significant. In this case, the t value is 3.67. Since 3.67 is greater than 2.101, the two groups are statistically different.

Multiple Choice

Which of these statements is correct? The t-test. . .

a)  Is the most elementary method for comparing two groups’ mean scores

b)  Assumes that the samples are not normally distributed.

c)  Assumes that the data are at the nominal level of measurement

d)  Does not use degrees of freedom

Concerning ANOVA which of these statements is correct.

a)  ANOVA can be used to simultaneously investigate several independent variables and two dependent variables.

b)  One-way ANOVA investigates two independent and one dependent variable

c)  A 2 X 2 ANOVA studies two independent variables, each with two levels

d)  ANOVA is essentially an extension of the Chi-square

Which of the following is not one of the required assumptions for ANOVA?

a)  Each sample is normally distributed

b)  The variances in each group are equal

c)  Nominal data

d)  The scores are statistically independent

Concerning basic correlational statistics which of these statements is correct.

a)  A positive relationship in correlation statistics exists when one variable increases while the other decreases

b)  The least used correlation statistic is the Pearson product-moment correlation

c)  Correlation is the same as cause and effect

d)  The degree of correlation varies between –1.00 and +1.00

Which of these statements is incorrect about multiple regression?

a)  Multiple regression is used to analyze the relationship between two or more independent variables and two dependent variable

b)  Multiple regression serves basically to predict the dependent variable using information derived from an analysis of the independent variables

c)  It is used by researchers to predict success in college, and sales levels

d)  The dependent variables in multiple regression are predicted by weighted linear combinations of independent variables

True/False

If the calculated chi-square value equals or exceeds the value found in the table, the differences in the observed frequencies are considered to be statistically significant. (T) In chi-square statistics, small samples may not produce significant results so that most researchers suggest that each category contain at least two observations. (F) ANOVA is essentially an extension of the Chi-Square goodness of fit. (F) Suppose the amount of time reading newspapers correlates with the amount of time watching television news. The correlation figure says nothing about the amount of time spent with each medium. (T) Multiple regression is parametric technique used to analyze the relationship between two or more independent variables and two or more dependent variables. (F)

Fill in the Blank

1. The parametric statistical methods usually used with (interval and ratio) data.

(Systematic variance) in data is attributable to a known factor that predictably increases or decreases all the scores it influences. In ANOVA, the scores from measurements are used to calculate a ratio of variance, known as the (F ratio). (Interaction) refers to the concomitant influence of two or more independent variables on the single dependent variable. In correlation, sometimes, the relationship between two variables is positive up to a point and then become inverse. When this happens, the relationship is said to be (curvilinear).

Short Answer