Sample Size in Multi-Year Content Analyses of Monthly Consumer Magazines

Abstract

This study examined efficient sampling for content analyzing monthly consumer magazines during a five-year period. As with other forms of media content, sampling of monthly consumer magazines is more efficient using a stratified sample than a simple random sample. When sampling from a five-year period, a constructed year of twelve issues (one issue randomly selected from among the issues in each month) is more efficient and accurate than twenty randomly selected issues - the smallest simple random sample meeting standards of efficiency on all content measures.

Evaluating Sampling Methods for Content Analysis of Twitter Data

“…Also, we examined both objective and subjective information variables. Previous studies tested only objective information variables such as news categories, length, size, as well as number of stories, photographs, and advertisements (e.g., Hester & Dougall, 2007; Lacy et al, 1998; Lacy, Riffe, Stoddard, Martin, & Chang, 2001; Riffe et al, 1993). To enhance generalizability, however, we included more sophisticated variables involving subjective characteristics such as a tone of candidates and election frames.…”

“…A substantial body of literature has tested the efficiency of different types of sampling methods with traditional mass media content. This includes daily and weekly newspapers, weekly and monthly news magazines, broadcast news, and online news (Hester & Dougall, 2007; Lacy, Riffe, & Randle, 1998; Lacy, Robinson, & Riffe, 1995; Riffe, Aust, & Lacy, 1993; Riffe, Lacy, & Drager, 1996; Riffe, Lacy, Nagovan, & Burkum, 1996). These studies have almost exclusively focused on periodical media outlets in which the volume and nature of media content are largely influenced by periodical variations in advertising and information gathering that affect space and time available for content (Riffe et al, 1993).…”

Gender and Race Representations of Scientists in Highlights for Children

“…Therefore, the sample universe equaled 756 magazines. In keeping with Lacy, Riffe, and Randle’s (1998) stratified sampling technique for monthly consumer magazines, 12 issues from 5 years worth of the magazines were randomly selected, giving each magazine between 1947 and 2010 an equal chance of being selected and yielding a total of 151 magazines.…”

The "I" of Embedded Reporting: An Analysis of CNN Coverage of the "Shock and Awe" Campaign

“…The Pearson correlation coefficients for the distances correlations were as follows: r = .995 for all categories coded for pronouns used by reporters, and r = .956 for all categories coded for pronouns used by others. Pearson correlation coefficients of .84 or better are considered to indicate good intercoder reliability (Lacy, Riffe, & Randle, 1998).…”