Compares 20 sets each of samples of four different sizes using simple random, constructed week, and consecutive day samples of newspaper content. Compares sample efficiency, and shows the superiority of constructed week sampling. (SR)

Lays out a systematic analysis of different experimental designs for message effects research. Discusses advantages of designs that include replications. Considers the issue of a fixed- versus random-effects analysis, and argues that replication factors should be treated as random except in very unusual circumstances. Examines in detail two…

Investigates the use of "vague quantifiers" (terms such as "often,""sometimes,""rarely," or "never") in communication research. Finds that these words do not always mean the same thing to different people, and thus may not constitute interval scales. Suggests that research outcomes based upon such…

Raises questions concerning the percentage of responses necessary to obtain accurate results from a survey. (RB)

Seeks to answer three questions: How successful are personal attempts to persuade others to change their opinions? How do conversations dealing with public affairs differ from conversations on other topics? and under what condition is information from the mass media cited in conversations? (RB)

Explores how to achieve rigor in applying methodology to research questions regarding quantitative approaches, specifically regarding statistical analysis, measurement, and sample decisions. (SR)

Finds that, although call-in poll participants are generally similar in socio-demographics to the population as a whole, they are more opinionated than are telephone survey respondents. Confirms that call-in polls and surveys of the population do not necessarily produce similar information concerning public opinion. (SR)

Describes a method for determining the accuracy of science news coverage. (RB)

Examines current practices in mass communication research with respect to message experiment design, analysis, and interference. Provides a logically consistent framework for mass communication researchers to make experimental design and data analysis decisions. (PRA)

Suggests avenues to close the schism between research methodology and theory in mass communication. (RB)

Proves the influence of three variables on election outcomes and demonstrates the potential utility of the panel design for detecting causality in analyses of political data. (RB)

Compares coverage of five subjects in eight daily newspapers, employing both factor analysis and Pearson correlations. Computes correlation coefficients among newspapers by using (1) size of individual stories as items; and (2) space measurements by categories. Finds that different ways of coding the same raw data produced different results. (SR)