Published: July 07, 1997
Last Updated: October 16, 1997
Printer-friendly version

To take this comprehensive and complex snapshot of the American newsroom, in August 1996 the ASNE staff carefully designed a sample of newsroom employees. From a list of the 1,487 newspapers in the United States, the staff randomly selected 61 papers. A scientifically valid method known as ''stratified probability sampling'' was used to ensure proportionate representation from papers of all sizes. For example, three papers were randomly chosen from the over-500,000 circulation category, seven papers from the 250,000-500,000 group, etc., to reflect the actual distribution of newspapers of all sizes. The chart below summarizes the result. A small number of the newspapers in the original group of 61 declined to participate, either because of lack of interest, unwillingness to commit staff time, or policies precluding participation in surveys. In these instances, ASNE randomly chose a substitute newspaper from the same circulation group. Then, in late August, the directing editor of each selected paper was asked to randomly select a sample of journalists from his or her staff. Only full-time employees who fit the categories of reporter, copy editor, photographer/artist, supervisor or editorial writer were to be included. The editor was instructed to use an alphabetized roster and select every ''nth'' name. The number of participating employees depended on the size of the paper, again in order to maintain proportionate representation. For example, the over-500,000 papers were each asked to select 40 participants; the under-5,000 papers were each asked to select three staff members. Collection of the Data Each employee chosen to participate received a survey packet, which included the questionnaire with instructions and a computer-scannable answer sheet. Participants were assured, both in the written instructions and by the on-site survey coordinator, that responses would be entirely confidential. If an employee declined to participate, a substitute was to be chosen by use of the same ''nth name'' random process. Editors were asked to give the employees three days in which to complete and return the questionnaire. (In a pre-test with journalists in August, the questionnaires took an average of 30 minutes to complete.) Responses were returned to ASNE beginning in mid-September, and all the participating newspapers had returned their questionnaires by Oct. 15. Of the 61 newspapers eventually contacted, only three failed to return any questionnaires. All three were below 25,000 in circulation. Thus of the 1,191 questionnaires sent out, 1,037 were returned. This represents a response rate of 87.1 percent, which is considered unusually high in survey research. On Oct. 15 the answer sheets were sent to the Bureau of Evaluative Studies and Testing at Indiana University, which scanned the answer sheets, processed the data, and transferred it to the author (at the Indiana School of Journalism). The author used SPSS-Windows for the statistical analysis of the data. An Oversample for Better Minority Representation In order to get a sizable sample of minority journalists (to enable statistically valid conclusions), the ASNE staff embarked on a second phase of the study. Directors of the principal minority-journalist organizations (National Association of Black Journalists, National Association of Hispanic Journalists, Asian American Journalists Association, Native American Journalists Association, and National Lesbian and Gay Journalists Association) were asked to cooperate with ASNE in a mailing to their members, and they readily agreed. Instead of using a random-selection process, ASNE staff simply sent questionnaires to all members of the organizations who were employed at daily newspapers. Of these, 385 questionnaires were returned. (The responses undoubtedly would have been more numerous had not a glitch occurred in the mailing. Because of an unexpected delay in the delivery of survey forms to ASNE, and another delay in the delivery to the journalists by the Postal Service, many of the minority and gay journalists received the survey after the published deadline for returning it.) The ''oversample'' questionnaires were not added to the workforce sample in the statistical analysis, as the sampling methods for the two groups were different. However, the minority oversample questionnaires and the minority questionnaires from the workforce sample were combined for separate analysis, as were the gay/lesbian questionnaires from both sources. From the two sources, a total of 356 minority-group and 181 gay/lesbian questionnaires were analyzed. Because the number of American Indians was still too low (12) to assure statistical significance, the American Indian responses were not broken out as a separate ethnic category in the analysis. The questionnaire and the responses can be found at the back of this report. Wherever appropriate, the results from the 1996 survey are presented with the equivalent 1988 results. Participants also were invited to attach their own comments - either reactions to the questions or elaborations on their numerical answers - to the survey sheet. Forty-four journalists responded in this additional manner, and many of their comments were quoted in this report. Questionnaire On the next 12 pages are the responses to each question on the ASNE Workforce Survey questionnaire, broken down by gender, ethnic group and sexual orientation. The figures represent percentages of the group shown at the top of the column. The percentages originally totaled 100 for each column of each table. Because the figures shown here have been rounded to whole numbers, some columns may total 99 or 101. The key to the data in the columns: T = Percentages of the total responses from the original workforce sample (n = 1,037). '88 = Overall percentages from the 1988 ASNE workforce survey (n = 1,200). F = Female journalists in the workforce sample (n = 379). W = Whites in the workforce sample (n = 903). B = Blacks in the workforce sample and further survey of black journalists (n = 154). H = Hispanics in the sample and further survey of Hispanic journalists (n = 92). AA = Asian Americans in the sample and further survey of Asian American journalists (n = 98). G = Gays, lesbians and bisexuals from the sample and further survey of gay/lesbian/bisexual journalists (n = 181). Composition of the Sample  Circulation Category  Totral US employees  Total US papers  Percent of Workforce  Papers in sample  Responses in sample  Percent of sample  500,000 and over  5,964  13  10.8%  3  101  9.7%  250,000-500,000  8,362  29  15.2%  7  187  18.0%  100,000-250,000  11,228  72  20.4%  10  260  25.1%  50,000-100,000  8,853  119  16.1%  9  201  19.4%  25,000-50,000  8,159  219  14.8%  7  119  11.5%  10,000-25,000  7,948  441  14.4%  11  122  11.8%  5,000-10,000  3,346  373  6.1%  7  30  2.9%  5,000 and over  1,280  221  2.3%  7  17  1.6% Click here to continue with survey  | Back to Top | Table of Contents |