Exam 2 Comm 2300

Likert-type scales:

-are a common response set for closed questions. -assign numerical values to each response choice. -are interval level data. -can contain a neutral middle response. =ALL THE ABOVE

Content analysis can be performed on texts or messages that are:

-captured in writing -captured on audiotape -captured on videotape = ALL BUT D

Interaction analysis codes:

-communication into categories -the verbal or nonverbal features or functions of conversation -communication according to a standardized coding scheme = ALL THE ABOVE

The mean:

-is the most commonly reported measure of central tendency - is the most sensitive to extremely high or low scores. -depends on each and every score. -is also known as the average. = ALL THE ABOVE

Regression is superior to correlation because it:

-it is more flexible -can predict some variables by knowing others -can be used with more than two variables -can test for the influence of more than one independent variable = ALL THE ABOVE

Hypothesis testing:

-relies on sampling and significance levels. -is really decision making about accepting the alternative explanation or retaining the null hypothesis. -allows for Type I and Type II error. -is used in research projects based on quantitative methods. = ALL OF THE ABOVE

The t-test is:

-restricted to one independent variable -restricted to one dependent variable -restricted to two groupings or categories of the independent variable. -can test independent or dependent groups. = ALL THE ABOVE

In presenting the results for a statistical test of difference, the researcher should report the:

-statistical test -level of probability or significance achieved. -degrees of freedom for the test - = ALL BUT D

In presenting the results for a statistical test of difference, the researcher should report the

-statistical test -level of probability or significance achieved. -degrees of freedom for the test. -the null hypothesis = ALL BUT D

In content coding, unit of analysis refers to:

-the discrete thing that is coded and counted -the standard or uniform unit that allows comparisons among categories. = A and B

When a researcher uses an ANOVA, you can assume that:

-third choice: The independent variable is nominal level data and the dependent variable is continuous level data.

Exogenous variables are

-used in structural equation modeling (SEM) -variables that are not caused by another variable = BOTH A and C

Match the following terms and descriptions: 1. Information about the relationships between or among variables. 2. All participants' responses compiled together. 3. Participants' numerical responses to questionnaire. 4. Summary information about the data set as a whole. 5. Information about communication phenomena, of which numbers are one type.

1. Inferential statistics 2. Dataset 3. Raw data 4. Descriptive statistics 5. Data

Matching: 1. Describe how the majority of participants responded. 2. Numbers of sources from which data were collected. 3. Describes the spread of scores

1. Measures of central tendency 2. Number of cases 3. Measures of dispersion

A frequency is used to:

= C and D -report on the occurrence of communication events -compute percentages

A good survey item is one that is:

A and B. Reliable and valid; straightforward

The normal curve is:

A theoretical distribution of scores.

Behavior coding pretesting

A third person monitors the researcher asking questions of the participant.

In reading a quantitative research article, a consumer should:

ALL THE ABOVE

Degrees of freedom:

Are a way in which researchers account for variation due to error.

Correlations can also be computed for variables based on ordinal or nominal data. These test include:

B and C

A skewed curve is:

Chacterized by the data being bunched to one side or the other.

The chi-square:

Compares the observed frequency with the expected frequency.

Researchers often include stimulus statements with questionnaires or survey. The purpose of the stimulus statement is to:

Direct participants' attention to the type of questions they will be answering, as well as provide general instructions in answering the question.

Inferential statistics are used to:

Draw conclusions about a population by examine the data from a sample.

A negatively skewed curve represents a distribution in which there are very few scores on the right side of the distribution.

False

A significant r value would cause the researcher to retain the null hypothesis

False

Achieving statistical significance indicates that the findings have social and practical application.

False

Because there is no theoretical limit to the number of predictor variables tested in multiple regression, it is common for researchers to include 10 or more.

False

Calculating and identifying an acceptable level of intercoder reliability is fairly straightforward.

False

Chi-squares are limited to testing differences between two levels of any variable.

False

Content analysis is especially fruitful when researchers content analyze texts or messages without first specifying hypotheses or research questions.

False

Correlation equals causation.

False

Correlation tests for curvilinear relationships

False

In considering the distribution of scores, there are four standards to the right of the midpoint, and four standards to the left of the midpoint.

False

In content analysis, the frequency with which elements occurs is always a direct reflection of the element's value.

False

In tests of difference, any difference found will be statistically significant.

False

Inferential statistics tell the researcher which participant's score is the most signficant.

False

Inferential statistics tell the researcher which participant's score is the most significant.

False

Interaction analysis is best used to code the messages of an individual giving a speech.

False

Intercoder reliability is calculated for either unitizing reliability or for categorizing reliability as calculating reliability for both decisions in content analysis is unnecessary.

False

It is okay to include some items in a questionnaire for which are unsure that participants will have adequate knowledge to respond, if this is the case, participants will know to ignore this item and go on to the next.

False

Most distribution of scores have only one score that occurs most often.

False

Only empirical research reports are included in the literature review.

False

Regardless of the researcher's level of experience, it is always wiser to develop a new and unique questionnaire or scale rather than modifying or adapting existing questionnaires or scales.

False

Researchers use ANOVA when they are interested in testing for the variance that exists within groupings of the independent variable.

False

Researchers use ANOVA when they are interested in testing for the variance that exists within groups of the independent variable.

False

Response rate is the same as sample size.

False

Significance levels are never set at levels other than .05.

False

Statistical differences can be determined by simply comparing mean scores. If the mean scores are different, then the different is statistically significant.

False

Statistical test for relationship are limited to two variables.

False

Statistically significant results ensure practical application of the results.

False

Surveys on the same topic written to achieve different objectives will be composed of the same questions and response choices, and be subject to the same interpretations.

False

The brief conclusion and description of statistical tests in the results section eliminates the need to interpret results in the discussion section.

False

The choices in a response set to closed questions should be exhaustive and mutually exclusive, but do not need to be equivalent.

False

The problem statement is the same as and can act as a substitute for hypotheses and research questions.

False

The t-statistic is found by simply computing the means and standard deviations for both groups.

False

Training coders for an interaction analysis research project will eliminate all differences in their application of the coding scheme.

False

When writing the research report, it is sufficient to rely upon the spell checker in your word processing program to catch all spelling errors.

False

Conventional pretesting

Helps uncover questions which can be interpreted in multiple ways.

The standard deviation is the:

Indication of how close or how far apart scores are from one another.

Cognitive pretesting

Individuals like the persons who will be selected for the research project complete the survey or questionnaire just as participants will in the research project.

A semantic differential scale:

Is based on two bipolar adjectives.

Choose the term that matches the section of a quantitative research report being described below conclusions of statistical tests are presented without interpretation.

Literature Review

THE further out a score is on the distribution; the:

More extreme the score.

Internal reliability is the degree to which:

Participants like the questions asked on a survey.

Choose the term that matches the section of a quantitative research report being described below. Reporting accuracy is critical in this section.

Results section

A significance level is:

The level of error the researcher is will to accept for each statistical test.

Closed questions are most effectively used when:

The potential responses are known in advance

Recall cues are used:

To direct participants' attention to the issue in which the researcher is interested.

A chi-square can indicate if significant differences are found, but can not determine where the significance differences occurred.

True

A correlation coefficient can range from -1.00 to +1.00.

True

All surveys and questionnaires should have explicit directions for participants, telling them how and where to mark their responses.

True

An alternative to a self-report questionnaire is to ask the same questions in a one-on-one interview.

True

Categories used in content analysis can be derived from the theory or previous research, or can emerge from the data.

True

Content analysis is one type of data reduction technique. Regardless of the size of text or the number of messages, meaning can be classified into fewer categories.

True

Descriptive statistics supply information about the variables, but the researcher must make value judgments or inferences about what the data mean.

True

For ANOVA to be statistically signficant, the between-groups variance must be greater than the within-groups variance.

True

For ANOVA to be statistically significant, the between-groups variance must be greater than the within-groups variance.

True

Generally, it is not meaningful to analyze or interpret a score on one variable from one individual.

True

HAVING specified and systematic rules and procedures for selecting and coding content in content analysis helps to decrease coders' subjective analysis.

True

If the test of difference is significant, the research hypothesis is accepted.

True

In interpreting an ANOVA, the main effects are ignored if a significant interaction effect occurs.

True

It is possible that the difference the researcher predicts in the hypothesis is not the difference that will exist in the data.

True

MULTIPLE regression allows the researcher to determine the relative importance of each variable to the regression relationship.

True

Mean, median, and mode are measures of central tendency.

True

Regardless of how the research hypothesis is stated, there is a complimentary null hypothesis.

True

Regression is a particular advantage to communication researchers because it allows researchers to study variables that cannot be experimentally manipulated.

True

Researchers use inferential statistics to determine if the relationship observed in the data is stronger than the relationship that might occur due to chance.

True

Statistical tests were designed to test for relationships and differences on normally distributed variables.

True

THE null hypothesis is retained until there is sufficient statistical support for accepting the research hypothesis.

True

Tests for relationships are most commonly computed for variables of continuous level data.

True

The Reference list must be complete and without error because it is the mechanism by which readers track backward from the research report to find other research reports.

True

The inference that a signficant result for a sample will also hold true for the population is based on the principle of probability.

True

The inference that a significant result for a sample will also hold true for the population is based on the principle of probability.

True

The letter n or N represents the number of cases for which data is reported.

True

The null hypothesis is retained until there is sufficient statistical support for accepting the research hypothesis.

True

The significance level is usually set at .05 as the criterion for making the decision to accept the research hypothesis.

True

To compute the median, the data must be arranged in order from lowest to highest score.

True

To use a correlation as a statistical test, each participant must have provided measurements on two separate variables.

True

Two distributions of scores can have the same mean, but different spreads of scores.

True

In content coding, it is common to see coding schemes with one category identified as "other". This reflects:

a failure of the coding scheme, indicating that the coding scheme is not as developed as it should be.

In content analysis, manifest content is the:

description of the characteristics of the content itself.

Researchers computed the coefficient of determination to

determine the amount of shared variance for a significant value

A correlation matrix is used to:

display how every variable is correlated with every other variable

Content analysis and interaction analysis are similar in that

each is a method for analyzing content of interaction.

A spurious correlation is:

one in which a third variable influences the nature of the relationship between the third variable tested.

If the data for the variables being tested are not normally distributed, the statistical test

provides only an approximation of the relationship.

In the results section, the minimum information needed to be presented for each hypothesis or question is:

statistical test used, the result of the test, and a written description of the statistical test as support for or rejection of the hypothesis, or connecting the result of the statistical test to the answering of the research question.

A correlation is the statistical test for

testing the linear relationship between two continuous level variables

After reading the method section of a quantiative research report, the consumer should know:

what the researcher did to collect the data