• Two factors emerge at the top of the list every time journalists of color are asked why they might leave the profession-lack of professional challenge and lack of opportunities for advancement. It appears, from this research review, that no amount of work to improve other areas of the job-e.g., the quality of the working environment, supervisory relationships, pay, etc.-will result in greater retention of journalists of color unless these two issues are adequately addressed.
• In all four studies in which it was offered as a possible reason for leaving journalism, "[lack of] professional challenge" was the top reason chosen by newspaper journalists of color.
• In all studies that permitted comparisons between white and nonwhite journalists, perceived lack of advancement opportunities proved far more salient to journalists of color than to white journalists as a likely reason for leaving (newspaper) journalism.
• In two studies that compared the perceptions of journalists of color with those of newsroom managers, journalists of color were far more likely than managers (by more than 35 percentage points) to identify lack of opportunities for advancement as a primary cause for minority journalists leaving the field.
• For journalists of color, compensation appears to play less of a role than do a number of other factors in making a decision about whether to stay in journalism or not. But this is likely due to the fact that they are more concerned about perceived obstacles to advancement and lack of professional challenge on the job.

All things being equal, journalists of color are as concerned about the issue of inadequate pay as whites. However, it is clear that pay increases alone will not keep journalists of color in the field.

• Characteristics of the work situation-e.g., working conditions, stress levels, work load-appear to be neither the most important nor the least important factors affecting the retention of journalists of color.
• Poor relationships with supervisors do not appear to be driving journalists of color away from newspaper journalism any more (or less) than they appear to be driving white journalists away from the field. This does not mean that supervisory relationships are not an important factor in retention; rather, it means that they do not tend to loom any larger for journalists of color than they do for whites.

Again, because journalists of color are more concerned about perceived obstacles to advancement and lack of professional challenge on the job, having positive relationships with supervisors may be a necessary, but not sufficient, condition for keeping them in the field.

• There appear to be some aspects of the job of journalism itself that may be more likely to influence journalists of color than whites to leave the field of (newspaper) journalism, such as "not being able to cover stories that interest [them]." This may be related to the fact (documented in several studies) that journalists of color were more likely than whites to have entered journalism out of a desire to "have an impact on society," "to influence public affairs," or "to help people."

This report reviews findings from 13 studies conducted since 1989 relevant to the topic of minority newspaper journalist retention. Studies were selected in consultation with the American Society of Newspaper Editors.

Efforts were made to obtain copies of all studies that involved quantitative surveys of newspaper journalists of color, beginning with the ground breaking 1989 ASNE study, "The Changing Face of the Newsroom." To our knowledge, this was the first research study to include a large enough sample of newspaper journalists of color to permit separate analyses of that group. To date, the 677 newspaper journalists of color surveyed in that study remains the largest group of such journalists ever surveyed. (Some subsequent studies have included larger samples of journalists of color in general, but not larger samples of newspaper journalists of color specifically.)

Research that focuses on issues of concern to minority journalists has been conducted steadily throughout the last decade. At about the same time the ASNE study was conducted, the National Association of Black Journalists released a study called "Halting the Exodus," indicating an awareness of the problem of minority journalist retention already at that early date. It focussed on the careers of successful African-American media managers, attempting to identify the factors in their careers that had led to professional advancement.

In 1990, Alexis S. Tan conducted a study for the Asian American Journalists Association, titled "Why Asian American Journalists Leave Journalism and Why They Stay." Notably, this study included interviews with 30 Asian Americans who had left the journalism profession to pursue other lines of work, as well as interviews with 265 current members of AAJA. The small size of the sample of former journalists, though, makes generalizations problematic.

Two studies were released in 1991. One, "The Newsroom Barometer: Job Satisfaction and the Impact of Racial Diversity at U.S. Daily Newspapers" by Ted Pease and J. Frazier Smith, in many ways, followed-up and amplified the 1989 ASNE study. The other, the Newspaper Association of America's report on "Employee Departure Patterns in the Newspaper Industry," was the first study commissioned to draw a demographic profile of newspaper employees who had left their jobs. An explicit motivation for the study was to determine whether "the turnover rate among minorities at newspapers may be higher than among non-minorities."

In 1994, NABJ issued another report, "Muted Voices: Frustration and Fear in the Newsroom," that compared the perceptions of African American journalists and news managers on issues related to professional advancement. The report revealed some staggering differences in perceptions between the two groups concerning opportunities for advancement.

In 1995, the NAA brought out a follow-up report to its 1991 study on employee departures, "Preserving Talent: A Study of Employee Departures in the Newspaper Industry," that went considerably beyond the earlier report by conducting in-depth surveys with more than 2,000 former full-time newspaper employees, including nearly 400 news/editorial employees. It included an extensive battery of questions regarding the factors that influenced employees' decisions to leave their former place of employment.

The year 1996 saw the release of two massive studies of American journalists-"The American Journalist in the 1990s: U.S. News People at the End of an Era" by David H. Weaver and G. Cleveland Wilhoit and "Winds of Change: Challenges Confronting Journalism Education" by Betty Medsger. The former was a follow-up to Weaver and Wilhoit's definitive study of the journalistic workforce in 1982-83 (itself a follow-up to the landmark 1971 study, "The News People," by John W.C. Johnstone, Edward J. Slawski and William W. Bowman, published in 1976). "Winds of Change" included a survey of over 1,000 "new" journalists, with one to eleven years of experience, that asked a number of questions relevant to the issue of retention.

In 1997, ASNE issued a follow-up study to its 1989 report on the journalistic workforce at newspapers, "The newspaper journalists of the '90s." It brought many of the 1989 findings up to date and added new lines of questioning as well.

In 1999, the International Women's Media Foundation released "Women Journalists of Color: Present Without Power," which compared the perceptions of women journalists of color and newsroom managers on issues regarding the newsroom climate and professional development opportunities for minority women.

Finally, in 2000, two more studies were released-the Freedom Forum report, "Newsroom Diversity: Meeting the Challenge" by Lawrence T. McGill, and "Perceptions of Minority Journalists at Unity '99 Regarding Minority Managers and Job Satisfaction" by Maggie Rivas-Rodriguez, Federico Subervi, Sharon Bramlett-Solomon and Don Heider.

A number of other reports were reviewed for this study, but were not included in the meta-analysis because they did not involve quantitative surveys of journalists of color. Some provided analyses of trends in the employment levels of different minority groups, such as the 1992 and 1993 NAHJ studies on "Hispanics in the News Media." Others offered more qualitative assessments regarding the situation of journalists of color, such as the 2001 NABJ report, "Voices of Anger, Cries of Concern: Some NABJ Views of the Retention Problem-and Some Solutions." As amplifications of the concerns expressed through the results of the quantitative surveys, such studies are recommended reading.

Organization of this report: This report first provides a methodological overview of the 13 studies reviewed in this meta-analysis. (Detailed information on the methodology of each study is included in Appendix 4.) This is followed by a summary of findings from these studies regarding the expressed likelihood of journalists of color remaining in (newspaper) journalism.

The main section of the report - "Retention Hypotheses Tested" - focuses on what these 13 studies had to say specifically on the issue of retention, namely, what factors would be most likely to influence whether journalists of color either stayed in or left (newspaper) journalism. Results were analyzed with respect to seven key job dimensions-compensation, advancement opportunities, professional growth/empowerment, job content, the work environment, interpersonal relations, and family considerations.

The report includes four appendices. The first provides a summary matrix that describes the basic characteristics of each study analyzed in this report, including the purpose of the study, its publication date, the data collection methods, the groups surveyed, the sampling frame and details on the number of white and minority journalists surveyed.

Appendix 2 is a summary matrix that lists the subject matter of all questions asked across the 13 studies relevant to the topics of minority journalist retention, recruitment and job satisfaction. In other words, it provides a sort of general index to the questions asked across the 13 studies.

Appendix 3 is a matrix that summarizes the results from the 13 studies that pertain specifically to the factors that affect the decisions of journalists either to stay in or to leave (newspaper) journalism. In effect, it summarizes in tabular form the discussion provided in the main section of the report, "Retention Hypotheses Tested."

Appendix 4 provides detailed information on the methodologies and limitations of each study analyzed in the report, as well as a list of the most important findings from each study. It serves as an important supplement to the main body of the report, in that it includes many findings relevant to the topic of minority journalist retention, recruitment and job satisfaction that are not specifically discussed in the body of the report.

About the studies

The 13 studies analyzed in this report fall into three categories:

1. Cross-sectional studies: six studies involving a representative cross-section of working journalists (four of which focussed specifically on newspaper journalists, two of which looked at journalists in general),
2. NAA newspaper departure studies: two Newspaper Association of America studies involving former newspaper employees
3. Studies of specific minority journalist populations: five studies focussing on specific minority journalist populations

The six cross-sectional studies are:

• 1989a: Changing Face of the Newsroom (ASNE Human Resources Committee)1991a: Newsroom Barometer (Pease and Smith)1996a: American Journalist in the 1990s (Weaver & Wilhoit)1996b: Winds of Change (Medsger)1997: Newspaper Journalists of the 90s (ASNE/Voakes)2000a: Newsroom Diversity: Meeting the Challenge (McGill)

The two NAA newspaper departure studies are:

• 1991b: Employee Departure Patterns (NAA)
• 1995: Preserving Talent (NAA/Watson Wyatt Worldwide)

The five studies of specific minority journalist populations are:

• 1989b: Halting the Exodus (NABJ)
• 1990: Why Asian American Journalists Leave Journalism (Tan)
• 1994: Muted Voices (NABJ)
• 1999: Women Journalists of Color (IWMF)
• 2000b: Perceptions of Minority Journalists at Unity '99 (Rivas-Rodriguez, et. al.)

Details on the research methodologies and a summary of key findings from each study may be found in Appendix 4.

What follows in this section of the report is an overview of the research designs and limitations of the studies in each of these three categories. In general, the following caveats should be kept in mind when interpreting the results presented in this report (each of which is explained in greater detail below):

• Populations sampled: Some studies focussed strictly upon newspaper journalists (1989a, 1991a, 1997, 2000), while others included journalists from both print and broadcast news organizations. The NAA departure studies (1991b, 1995) included non-news/editorial personnel as well as journalists.
• Sampling methodology: Most of the studies followed standard scientific sampling procedures and achieved reasonably high response rates, yielding results that are generalizable to particular populations within specified margins of error. Caution is advised, however, in generalizing from the results of two studies: 1999 (low response rates) and 2000b (possibly non-representative sample).
• Sample sizes: Many of the studies do not include large enough sample sizes of specific racial/ethnic groups to permit reliable generalizations to their respective populations. For example, only one study (1996a) included a large enough sample of Native American journalists to permit some degree of generalization to that group.
• Journalist of color samples: All of the journalist of color samples consist either mostly or completely of members of the four minority journalist associations-NABJ, NAHJ, AAJA and NAJA; the results from these samples are technically generalizable only to the membership of those organizations and not to minority journalists in general.

Additional caveats apply to aspects of particular studies, as well (as noted below).

Cross-sectional studies: Of the six surveys involving a representative cross-section of working journalists, three were conducted by mail, three by phone. All six studies used stratified samples to achieve proportionate representation of newspaper journalists at different size papers. In four studies, the sampling frame included all journalists at daily newspapers of all sizes. In the McGill study, only journalists at papers with circulations above 25,000 were included for analysis. In the Medsger study, only journalists with one to 11 years of experience were included for analysis.

All six studies relied upon oversamples of minority journalists, based on membership lists of the four minority journalist associations, ranging from an oversample of 210 journalists in the Pease and Smith study to 599 in the 1989 ASNE study. The McGill study, however, did not mix the sample of journalists of color from the minority journalist associations with the random sample of minority journalists obtained in the cross-sectional study, preferring to generalize unambiguously to members of minority journalist associations, rather than with some uncertainty to all journalists of color.

Three of the four cross-sectional studies of newspaper journalists included more than 800 white journalists in their cross-sectional samples (the McGill study included 349). In the Weaver & Wilhoit study, newspaper journalists made up 55 percent of the sample (yielding about 570 white newspaper journalists), while in the Medsger study, newspaper journalists made up about 53 percent of the sample (yielding about 450 white newspaper journalists). The margins of sampling error associated with these studies (for the white newspaper journalist samples specifically) range from +/- 6 percentage points for the McGill study to +/- 3 percentage points for the 1989 ASNE study.

The journalist of color samples (including both newspaper and non-newspaper journalists) in these six studies ranged from 341 in the Weaver & Wilhoit study to 677 in the 1989 ASNE study. To the extent that "journalists of color (at all news organizations)" is defined as the group of interest to be analyzed, then all six studies include a sufficient number of journalists of color to permit generalization within a margin of error of about +/- 6 percentage points. If "newspaper journalists of color" is defined as the group of interest to be analyzed, then the margins of error in both the Weaver & Wilhoit and Medsger studies increase to about +/- 7.5 percentage points.

For purposes of this report, which summarizes a large amount of data from many studies, it will be useful to focus most of the discussion on (newspaper) journalists of color in general. But it should be kept in mind that important differences exist among racial and ethnic groups that should not be overlooked in applying the findings in this report.

The sample sizes for specific racial/ethnic groups of newspaper journalists of color were too small in both the Weaver & Wilhoit and Medsger studies to permit reliable generalization. In the other studies, the smallest sample sizes for specific groups of newspaper journalists of color (except Native Americans) are found in the 1997 ASNE study-154 African Americans, 92 Hispanics/Latinos and 98 Asian Americans. The margins of error associated with the two smallest samples are probably +/- 7 percentage points or larger (depending upon the size of the populations being generalized to). For the African American sample, the margin of error is, at minimum, about +/- 5.5 percentage points. The margins of error associated with specific racial/ethnic groups in the other studies would be slightly smaller.

In general, when looking for differences between the responses of different racial/ethnic groups in these studies, a gap of less than 10 percentage points between the responses is not likely to be statistically significant. This doesn't necessarily mean that the gap doesn't exist in reality-but it does mean that we can't be more than about 90 percent certain that the gap exists.

Unfortunately, there have not been enough Native American newspaper journalists included in any of the studies to permit reliable generalizations to that group. Only in the Weaver & Wilhoit study were there enough Native American journalists (n = 84) to permit some level of analysis. However, this group included both print and non-print journalists.

It should be kept in mind that generalizations to all newspaper journalists of a specific racial or ethnic group are somewhat compromised by the fact that the samples analyzed in each of these studies consisted largely of members of the four minority journalist associations (NABJ, NAHJ, AAJA, NAJA). They do not represent a truly random sample of newspaper journalists of color. Members of minority journalist associations make up from 47 percent to 100 percent of the journalist of color samples in these six studies.

This may or may not be a problem. It all depends on how representative the newspaper membership of minority journalist associations is of minority newspaper journalists in general. At present, this must be considered an open question.

In 1992, Weaver and Wilhoit compared the characteristics of a small sample of minority journalists drawn from a random sample of journalists (43 African Americans, 25 Hispanics, 12 Asian Americans and 7 Native Americans) with the characteristics of a sample of minority journalists drawn from the membership lists of the four minority journalist associations (34 African Americans, 55 Hispanics, 88 Asian Americans and 77 Native Americans). Finding more similarities than differences between the paired samples of minority groups across 15 variables, they opted to combine the paired samples with each other. Given the small sample sizes involved for each of these groups, this analysis cannot be considered conclusive, however.

Nevertheless, most researchers have opted to combine the oversamples of minority journalist association members with the journalists of color obtained in random samples, including ASNE (1989), Pease and Smith (1991), ASNE (1997), Weaver & Wilhoit (1996) and Medsger (1996). McGill (2000) chose not to combine the two types of minority samples.

Regardless of whether minority samples have been combined or not, it should be kept in mind that, in all studies conducted to date, what we know about the attitudes and experiences of minority journalists is based primarily upon the responses of members of the four minority journalist associations.

NAA newspaper departure studies: The two NAA newspaper departure studies were both conducted by mail. The first (1991b) collected basic demographic data on departing newspaper employees through a survey sent to publishers at 371 newspapers. The second (1995) collected a substantial amount of information directly from departed employees, representing 295 newspapers.

Response rates in both studies were fairly low (24 percent and 17 percent, respectively), although the sampling frames for both were very large, resulting in (presumably) large sample sizes in both studies. (The actual sample size was not provided in the earlier study, but the number of departing employees is inferred to be over 2,000.) The low response rates could mean that the samples are less representative than they ought to be, but it is not clear in which direction, if any, the samples may be biased.

There are several important limitations to the findings from these two studies, insofar as the results may pertain to the issue of retaining newspaper journalists of color:

• Most of the findings pertain to newspaper employees in general, rather than to newspaper journalists specifically.
• The findings include employees who took new jobs within the newspaper industry, as well as those who left the industry altogether.
• The sample in the second NAA study probably includes fewer than 60 newspaper journalists of color who left the industry, too small to permit reliable generalizations. (The first study did not attempt to do anything other than describe the demographic characteristics of departing employees, making the size of the newspaper journalist of color sample irrelevant, except insofar as it yielded an estimate of the proportion of the total sample from different racial/ethnic backgrounds.)

Nevertheless, with one exception (the 1990 Tan study of Asian American journalists, which included only 30 departed journalists), these are the only studies available that describe the characteristics of newspaper journalists who have actually departed from the industry.

Studies of specific minority populations: Over the past decade, separate studies have focussed on African American journalists (NABJ, 1993), Asian American journalists (Tan, 1990), women journalists of color (IWMF, 1999), senior black media managers (NABJ, 1989b) and journalists of color in general (Rivas-Rodriguez, et. al., 2000).

In all of these surveys, the focus was on journalists in general, not newspaper journalists per se. However, as in other studies involving journalists in general, newspaper journalists typically comprised a small majority of the samples, ranging from 55 percent in the Tan study to 63 percent in the Rivas-Rodriguez, et. al., study.

Three of the five studies involving specific minority journalist populations were conducted primarily by mail. A fourth study (IWMF, 1999) was conducted primarily online, while the fifth involved in-person intercepts coupled with self-administered questionnaires (Rivas-Rodriguez, et. al., 2000).

The IWMF and Rivas-Rodriguez, et. al., studies departed the furthest from orthodox survey research practices, making the results in these studies less reliable than those obtained in the other three. The IWMF study reported very low response rates-15 percent among women journalists of color and 7 percent among "newsroom managers," although the total number of completed surveys among each group is on a par with the sample sizes obtained in most of the other studies reviewed in this meta-analysis. In self-selected samples, however (such as in mail and online surveys), such low response rates are troublesome.

The Rivas-Rodriguez, et. al., study basically made use of a convenience sample (i.e., attendees at the 1999 Unity Convention), which typically does not allow the results to be generalized to a larger population (i.e., journalists of color in general). However, the study did manage to collect data from a significant proportion of all Unity '99 attendees, which should have had a stabilizing effect upon the results obtained. Still, it is not possible to generalize from the results of this study to a broader group than journalists of color who attended Unity '99.

Methodologically, the most solid study of the five was the 1993 NABJ "Muted Voices" study. It achieved a 67 percent response rate from a mail survey of a random sample of 800 members of the National Association of Black Journalists. It also drew a comparative sample of 100 news managers, but information about the response rate for that sample was not provided.

The 1990 Tan study of Asian American journalists did not achieve quite as high a response rate (38 percent). For comparative purposes, it did include some interviews with Asian American journalists who had left journalism, but the sample size was too small (n = 30) to permit meaningful generalizations.

The 1989 NABJ study of senior black media managers collected data from 36 individuals, from a list of 70 identified by NABJ, for a 51 percent response rate. If the true population of "senior black media managers" at the time was just 70, then the responses of these 36 should provide a fairly accurate portrayal of that group. It is likely, though, that the total number of people in this group was larger than 70, although how much larger is difficult to gauge.

It is possible to estimate margins of sampling error for the "Muted Voices" and Tan studies, of +/- 4 and 6 percentage points respectively. Due to the methodological limitations of the IWMF and Rivas-Rodriguez, et. al., studies, it is not possible to calculate specific margins of error for those studies. The findings in those studies must be considered suggestive only.

Likelihood of Staying in (Newspaper) Journalism

Trends in both the annual ASNE newsroom census numbers and in studies over the past decade suggest that the problem of retaining newspaper journalists of color is getting worse. In 2001, the ASNE newsroom census showed, for the first time, an actual decline in the percentage of newspaper journalists of color in the workforce, meaning that more journalists of color chose to leave the newspaper business in 2000 than were hired.

In 1989, ASNE reported in its "Changing Face of the Newsroom" study that journalists of color were only slightly less likely than whites to say they planned to make a career out of journalism-64 percent of African Americans, 70 percent of Asian Americans and 69 percent of Hispanics, compared to 74 percent of whites. (Although this question asked about "journalism," rather than "newspaper journalism," it is likely that for most respondents "newspaper journalism" was implied.)

In 1991, Pease and Smith reported that "journalists of color …are no more likely than their white co-workers to be thinking about leaving the profession in the next several years: 18.5 percent of minority journalists say they are unlikely or very unlikely to be in newspapers in five years, but so do 14.3 percent of white journalists."

In 2000, however, McGill reported that journalists of color were much less likely to say they planned to make a career out of newspaper journalism-39 percent of African Americans said they planned to stay in newspaper journalism, compared to 43 percent of Asian Americans, 50 percent of Hispanics and 66 percent of whites. Further, nearly one in four journalists of color (23 percent) said they might leave the newspaper business within the next five years.

Studies of specific minority journalist populations bear out the finding that many journalists of color do not expect to remain in journalism. Findings related to African American journalists are especially consistent across studies; this is less true for Hispanics and Asian Americans. The only study that included enough Native American journalists to permit analysis (Weaver & Wilhoit, 1996) showed that Native American journalists were also at high risk for leaving the profession.

• The 1993 NABJ study found that one in four NABJ members (24 percent) said they planned to stay in journalism for five years or less.
• The 1999 IWMF study found that just two-thirds of women journalists of color (66 percent) saw "a future for themselves in the media."
• In 1990, Tan found that more than one in three Asian American journalists (36 percent said they were either "very likely" or "likely" to leave journalism in the next five years. (More than half of Tan's sample, 55 percent, were newspaper journalists.)
• The 1996 Weaver & Wilhoit study found Native American journalists to be the most likely of any racial/ethnic group to say they expect to be working somewhere else besides the news media in 5 years-29 percent, compared to 26 percent of African Americans, 19 percent of Hispanics, 11 percent of Asian Americans and 21 percent of whites.

Especially alarming is Medsger's 1996 finding that among "new" journalists, with one to 11 years of experience, journalists of color are already more likely than whites to say they are considering leaving journalism-48 percent to 42 percent.

Retention Hypotheses Tested

Each of the studies included in this meta-analysis was selected because it addressed, either directly or indirectly, the issue of retaining minority newspaper journalists. All but one study (NABJ, 1989b), asked specific questions concerning possible reasons for staying at or leaving one's job as a journalist.

The questions related to retention in these studies can be divided into seven broad areas:

• Compensation
• Advancement opportunities
• Professional growth/empowerment
• Job content
• Work environment
• Interpersonal relations, and
• Family considerations

Based on the frequency with which different types of questions were asked across these studies, it is clear that most researchers in this area assume that compensation and advancement opportunities factor significantly in decisions about whether or not to stay in (newspaper) journalism. Nine studies asked about the role of compensation in this decision, while 10 attempted to assess the role of perceived advancement opportunities.

Six studies looked at how issues of professional growth or "empowerment" might factor into the decision to stay in or leave journalism-e.g., professional challenge, involvement in decision-making, amount of autonomy, organizational openness to suggestions about work, the chance to develop a specialty, and training and development opportunities.

Four studies included questions concerning the content of the job-e.g., boring assignments, not being able to cover stories that interest you, dissatisfaction over editorial policies, and general dissatisfaction with journalism.

Six studies also asked questions about aspects of the work environment that might cause one to leave journalism-e.g., poor general working conditions, stress on the job, long hours and burnout.

Four studies examined various aspects of interpersonal relations as they related specifically to the decision to stay in or leave journalism. The most important relationship, as has been demonstrated in many studies across industries, is the one between the employee and his or her immediate supervisor; a number of studies went into this in some detail. Additionally, studies have looked at the role played by possible factors such as difficulties with management in general, personality conflicts, cultural conflicts, lack of recognition and feeling isolated from colleagues.

Finally, several studies attempted to assess the role played by family considerations in the decision to stay in or leave one's job.

All of these are factors that may have an impact on whether someone, of whatever race, remains in journalism or not. So, the question that needs to be answered is whether for members of minority groups some of these factors are more likely to play a significant role than they do for white journalists.

Compensation: There has long been a great deal of discussion about the need to increase pay levels in journalism-indeed, as Betty Medsger reported in 1996, students who go into journalism are the lowest paid of any college educated people entering the public or private work force. However, for journalists of color, compensation does not appear to play as big a role as do a number of other factors in making a decision about whether to stay in journalism or not. This is not to say that it plays no role. For many, it ranks among the top three or four reasons for leaving newspaper journalism.

In two of the studies of newspaper journalists, there appeared to be little difference between whites and nonwhites concerning the role of compensation as a factor in the decision to stay in or leave newspaper journalism. In the Pease & Smith (1991) study, for example, both whites and minorities regarded "professional challenge" as the most important of four factors that might affect their decision about staying in newspaper journalism, followed by "financial reasons." In the McGill (2000) study, "financial reasons" ranked 5^th out of 10 factors for whites, and 6^th for journalists of color.

In other studies of newspaper journalists, though, compensation emerged as a more salient factor for whites than it was for journalists of color. For example, it ranked first for whites among five factors listed in both the 1989 and 1997 ASNE studies. For journalists of color, "financial reasons" ranked third in the 1989 study, well behind "professional challenge" and "advancement opportunities." It ranked second in the 1997 study, again well behind "advancement opportunities" ("professional challenge" was not included as an option in the 1997 study).

The lower salience of pay for journalists of color (as compared to whites) also recurs in other studies of journalists in general. Weaver & Wilhoit (1996) found that, for journalists of color, "low pay" was the fourth most-often mentioned reason for wanting to "work outside the news media," but it was the top-mentioned reason among whites. When Medsger (1996) asked, "What is it that might make you want to leave journalism?" 34 percent of whites said low pay, compared to just 11 percent of journalists of color. And in the 1990 Tan study of Asian American journalists, low pay ranked 9^th out of 14 possible reasons for leaving journalism.

Since more than half of the journalists in these three studies were newspaper employees, these findings lend additional support to the notion that pay may be less important to journalists of color than it is to whites as an inducement for staying in journalism. To keep journalists of color at newspapers would seem to require something more than just higher salaries.

Advancement opportunities: In all studies that permitted comparisons between white and nonwhite journalists, the issue of advancement opportunities proved far more salient to journalists of color than to white journalists. Moreover, it consistently ranked at or near the top of the list of all factors that might influence the decision of a journalist of color to stay in newspaper journalism.

In three studies (ASNE 1989, Pease and Smith 1991, ASNE 1997), respondents were asked to choose the single factor from a list of four or five that would be "the most important factor in why you [might leave] the newspaper business." For white journalists, in all three studies, the issue of advancement opportunities ranked last among the reasons they might leave the field.

For newspaper journalists of color, however, advancement opportunities ranked second (by one percentage point) in the earlier ASNE study and first in the later ASNE study. In the Pease and Smith study, journalists of color ranked it third (behind "professional challenge" and "financial reasons"); still, they were far more likely than whites (17 percent vs. 6 percent) to say it was "very important."

In the McGill (2000) study, newspaper journalists were asked to evaluate how significant each of 10 factors might be in deciding whether to stay in newspaper journalism. Again, the contrast between journalists of color and whites on the issue of advancement opportunities was striking-61 percent of journalists of color said this would be a "major factor" in deciding whether or not to leave newspaper journalism, compared to 39 percent of whites. Among journalists of color, it ranked 2^nd only to "interest in another field" as a possible reason for leaving newspaper journalism. Among whites, it ranked 6^th.

Similarly, Weaver and Wilhoit (1996) found that journalists of color were much more likely than whites to say that "the chance to get ahead in the organization" was a "very important" factor in how they judged jobs in their field, 59 percent vs. 37 percent. Among journalists of color, it ranked 3^rd out of a list of nine items. Among whites, it ranked 6^th.

In the 1995 NAA departure study, news/editorial department employees who actually left newspaper jobs in 1994 cited "[lack of] opportunities for advancement" as one of the top three reasons (out of a list of 25) for leaving their jobs. Since this group included both whites and nonwhites, as well as individuals who went to other newspapers rather than leaving the industry altogether, it is not possible to draw any conclusions about the relative weight this factor played in the decisions of minority and non-minority journalists. It does suggest, however, that perceptions of advancement opportunities, along with perceptions of fairness in promotions and pay (the other two top reasons cited) are significant factors in the decisions of employees (of all races) to leave their jobs.

So central is assumption that advancement opportunities are key to minority journalist retention that the 1993 NABJ study focussed exclusively on that topic. In that study, 40 percent of African American journalists who expected to leave the field of journalism within five years indicated that "lack of promotion/career advancement opportunities" was a "very significant" factor in their decision to leave the business.

Even more importantly, the NABJ study revealed an amazingly wide gap between the perceptions of African American journalists and newsroom managers with respect to issues of advancement. For example, while nearly all of the newsroom managers surveyed (94 percent) said that "managers in [their] newsroom show a commitment to retaining and promoting black journalists," just 28 percent of NABJ members agreed with this.

This perceptual gap persisted across other items, as well. By margins exceeding 50 percentage points, African American journalists were more likely than newsroom managers to believe that:

• Blacks are less likely than other journalists in the newsroom to be considered for career opportunities (73 percent vs. 2 percent)
• Standards of promotion for blacks are higher than they are for other journalists in the newsroom (59 percent vs. 1 percent)
• Blacks spend more time in entry level positions (67 percent vs. 12 percent)
• Blacks are less likely to be taken under the wing of a more seasoned colleague (73 percent vs. 16 percent)

From the perspective of NABJ members, the five most serious problems affecting the advancement possibilities of African American journalists (each cited as either a "very" or "somewhat serious problem" by more than three-quarters of NABJ members) are:

• Too few blacks in jobs from which promotion candidates are selected
• Lack of commitment to diversity by supervisors
• Lack of mentors/role models
• Unrealistic perception of performance by managers, and
• Differences in culture/background

Newsroom managers also named four of these items as being among the top five problems affecting the advancement possibilities of African American journalists; each was cited as either a "very" or "somewhat serious problem" by a majority of managers. The predictable exception was "unrealistic perception of performance by managers" - 78 percent of NABJ members cited this as either a "very" or "somewhat serious problem," compared to just 24 percent of managers.

In its place, managers cited "lack of experience" as one of the top five problems affecting the advancement possibilities of African American journalists-71 percent of managers said this was either a "very" or "somewhat serious problem," compared to 41 percent of NABJ members.

These differences in perceptions between journalists of color and newsroom managers concerning advancement opportunities were also documented in the 1999 IMWF study of women journalists of color. More than half of the women journalists of color surveyed (54 percent) said that "obstacles to advancement" was one of the main reasons that women of color left their jobs. It was the second most frequently cited reason (out of a list of nine) given by women journalists of color (tied with "increase in pay").

Among newsroom managers, however, just 17 percent thought that "obstacles to advancement" played a role in the decisions of women journalists of color to leave their jobs. It ranked sixth among managers as a perceived reason that women of color left their jobs.

The most frequently cited reason given by women journalists of color for women of color leaving their jobs was "seeking new opportunities/challenges," a factor not unrelated to the perception of "obstacles to advancement" in one's current job. Interestingly, this was the top reason given by newsroom managers, as well. What this suggests is that newsroom managers tend not to see a relationship between lack of opportunity within their own organization and the seeking of new opportunities outside the organization, a connection all too readily made by women journalists of color.

The 1990 Tan study of Asian American journalists echoes these findings. In that study, the second most often given reason for possibly leaving journalism (from a list of 14) was "better opportunities in another field." Close behind, in third place, was "lack of advancement opportunities." (The top reason was "professional challenge," discussed below.)

Professional growth/empowerment: For both white and minority newspaper journalists, being professionally challenged on the job stands at or near the top of the list of factors that would cause them either to stay in or to leave newspaper journalism.

• In the 1989 ASNE study, "professional challenge" was the top reason given by newspaper journalists of color, from among five factors, for possibly leaving the field. It was the second most frequently chosen reason by white journalists (behind "financial reasons").
• In the 1991 Pease and Smith study, "professional challenge" was the top reason given by both newspaper journalists of color and white journalists, from among four factors, for possibly leaving the field.
• In the 1990 Tan study of Asian American journalists, "the need for other challenges" was the top reason given, from among 14 factors, for possibly leaving journalism.
• In the 1999 IWMF study of women journalists of color, "seeking new opportunities/ challenges" was the top reason cited, from a list of nine, for women of color leaving journalism.

It should be noted that "professional challenge" does not necessarily imply "freedom from supervision," which the Tan study (1990) showed to be a relatively unimportant factor in the decision to stay or leave. However, it may be related to such things as the relative "amount of autonomy" one has on the job and having the "chance to develop a speciality," both of which were found by Weaver and Wilhoit (1996) to be fairly important dimensions in how both white and minority journalists rated different jobs in their field.

Job content: Relatively few studies have focussed on the role that job content per se plays in whether or not a journalist stays in the profession. Most studies have focussed, instead, on the sociological aspects of the job-opportunities for advancement, professional growth and development, interpersonal relations, and the like. But for journalists of color, there appear to be some aspects of the job of journalism itself that may influence whether they stay in or leave the field.

In the McGill (2000) study, for example, 50 percent of newspaper journalists of color said that "not being able to cover stories that interest [me]" would be a "major factor" in their decision to either stay in the field or leave. Among whites, just a third (32 percent) felt this way. This may be related to the fact that newspaper journalists of color were also more likely than whites to say they were motivated to enter journalism by the opportunity to have an impact on society-two-thirds said this was very influential in their decision to become a journalist, compared to 39 percent of whites. The ability to make an impact on society depends, in no small measure, upon having the opportunity to cover stories that matter to journalists of color.

In an open-ended question, Weaver and Wilhoit (1996) asked journalists who said they would probably be working in some other field in five years why they thought they might leave journalism. Among journalists of color, the most frequently given answer was "disappointment with journalism-especially superficiality of news coverage, a tendency not to look at why stories were happening, narrow-mindedness and resistance to change." Answers of this type were given by about one in five journalists of color (20 percent), compared to 15 percent of journalists in general. (Among white journalists, as noted above, the most frequent answer was "low pay," 21 percent).

Weaver and Wilhoit also asked an interesting question about the various criteria by which journalists judge prospective jobs in their fields. Interestingly, in contrast to the mostly non-content-related reasons many journalists give for leaving the field, a substantial majority of both white and minority journalists (70 percent and 72 percent, respectively) said that a news organization's "editorial policies" were "very important" in evaluating a job at that organization. For whites, no other criterion was rated as more important. Among journalists of color, it was the second most frequently cited criterion.

Only "the chance to help people" (77 percent) was cited more often by journalists of color as a criterion for evaluating a job in journalism, echoing McGill's finding that most journalists of color said they were motivated to enter journalism by the chance to have an impact on society. In addition, Weaver and Wilhoit found that journalists of color were much more likely than whites to say that "the chance to influence public affairs" was "very important" to them in judging a prospective job in journalism (58 percent vs. 37 percent).

In other words, it appears that the content-related aspects of a job may be very important in choosing a particular job opportunity, while the sociological aspects of a job emerge as significant factors in determining whether an employee stays or leaves.

Work environment: Characteristics of the work situation-e.g., working conditions, stress levels, workload-appear to be neither the most important nor the least important factors affecting the retention of journalists of color. In most studies, they fall somewhere in the middle, except insofar as they may contribute to the psychological condition of "burnout," which does appear to play a fairly significant role in whether journalists remain in their jobs.

Only two studies, however, looked specifically at burnout as a factor affecting retention. McGill (2000) found "burnout" to be almost as significant a factor as "lack of advancement opportunities" in the decision of journalists of color to stay in or leave newspaper journalism. For both white and minority journalists, it ranked third among 10 possible factors.

Medsger's survey of new journalists (1996) found "burnout" to be a slightly more significant factor for minority journalists who were considering leaving journalism than it was for whites. In her survey of newsroom recruiters and supervisors, in the same study, "burnout" was cited by 70 percent of recruiters and supervisors as a "major" reason people (of all colors) left journalism, second only to "low pay" (72 percent).

Interpersonal relations: Interestingly, no studies have documented significant differences between white and minority journalists in terms of how they rate the quality of their relationships with their immediate supervisors. In other words, poor supervisory relationships per se do not appear to be driving journalists of color away from newspaper journalism any more (or less) than they appear to be driving white journalists away from the field. This does not mean that supervisory relationships are not an important factor in retention; rather, it means that they do not tend to loom any larger for journalists of color than they do for whites.

Some studies, in fact, document fairly strong positive relationships between journalists of color and their supervisors. In the McGill (2000) study, journalists of color were equally as likely as white journalists to say that their "immediate supervisor welcomes [their] ideas and suggestions about the newsroom" and that their "immediate supervisor advocates [their] story ideas to senior editors." They were only slightly less likely than whites to agree that their "immediate supervisor cares about [their] professional development as a journalist."

The 1995 NAA departure study examined, in greatest detail, the role of the supervisory relationship in relation to employee retention and found, as have studies in other industries, that the supervisory relationship plays a significant role in retention. Two aspects of the supervisory relationship-"equitable treatment by supervisor" and "contribution is valued by supervisor"-figured prominently in the decisions of journalists who left their jobs in 1994, although not quite as prominently as the issues of professional advancement and pay. Still, they ranked in a tie for 4^th out of 25 factors examined. Three other aspects of the supervisory relationship-"respectfulness of supervisor," "supervisor's management skill" and "supervisor's concern about employee's success" also ranked among the top 10 factors measured.

Beyond the supervisory relationship, a couple of studies have shown that other issues involving interpersonal relations do not appear to play as large a role in retention, although they do tend to play a somewhat larger role for journalists of color than for whites. For example, McGill (2000) found that journalists of color were more likely than whites to say that "feeling isolated from colleagues" was a factor that would affect their decision to stay in the field. Nevertheless, for both minority and non-minority journalists, it was still the lowest ranked of 10 factors examined in the study.

Similarly, the 1999 IWMF study found that women journalists of color were more likely than newsroom managers to cite such factors as "personality conflicts" (34 percent vs. 10 percent, respectively), "discrimination" (30 percent vs. 5 percent) and "cultural conflicts" (17 percent vs. 6 percent) as reasons that minority women had left journalism. But none of these were among the top reasons offered for the departure of these women.

Family considerations: In comparison to other factors, family considerations don't appear to weigh as heavily in the decisions of journalists of color to stay in or leave journalism. In the 1989 and 1997 ASNE studies and the 1991 Pease and Smith study, "family considerations" ranked last among possible reasons newspaper journalists of color might leave the field. (Among white journalists, it ranked next to last in all three studies, while "advancement opportunities" ranked last.)

In the McGill (2000) study, however, "family considerations" loomed larger as a factor related to the retention of minority newspaper journalists. Unlike the ASNE and Pease and Smith studies, respondents in the McGill study evaluated each item in a list of 10 possible reasons for leaving newspaper journalism, rather than choosing the one item from a list of four or five that would be the "most important factor" in deciding whether or not to stay in the field.

In the McGill study, 50 percent of newspaper journalists of color said that "family considerations" would be a major factor in their decision either to stay in or leave newspaper journalism. This was on a par with the percentage of journalists of color, for example, who said that "not being able to cover stories that interest you" would be a major factor in deciding whether or not to stay in the field. Both items were tied for 4^th among 10 possible reasons for leaving the field. For white journalists, "family considerations" ranked second only to "interest in another field of work" as a major reason for possibly leaving the field.

One conclusion that may be possible to draw from these seemingly discrepant findings across the four studies is that family considerations do matter as much for journalists of color as they do for white journalists. However, until journalists of color perceive that their opportunities for advancement are equivalent to those of white journalists, family considerations are not likely to be cited as one of the "most important" factors in deciding whether to stay in newspaper journalism.

Conclusions

Members of the National Association of Black Journalists and the Asian American Journalists Association have been surveyed nine times in the last 12 years for purposes of the 13 studies analyzed in this report. NAHJ members have been surveyed eight times, NAJA members six times. Altogether, an estimated total of nearly 4,800 journalists of color have been surveyed since 1989-2,382 African Americans, 1,230 Asian Americans, 916 Hispanics/Latinos and 270 Native Americans.

They have been asked repeatedly whether they expect to stay in journalism in general, or in newspaper journalism in particular. Across different surveys, between one-fifth and one-third have indicated that they do not expect to remain in journalism over the long term. Especially in more recent surveys, journalists of color have indicated a much stronger likelihood of leaving the field than have white journalists.

Two factors emerge at the top of the list every time journalists of color are asked why they might leave the profession-lack of professional challenge and lack of opportunities for advancement. This appears to be just as true for newspaper journalists of color as it is for journalists of color in general.

Inadequate pay is also a significant factor. But when asked to indicate "the most important reason" for possibly leaving the field, "professional challenge" and "opportunities for advancement" eclipse pay every time. Pay increases alone will not keep journalists of color in the field.

Clearly, a significant part of the problem has to do with perceptual differences between journalists of color and newsroom managers regarding opportunities for advancement. The 1993 NABJ study, "Muted Voices," demonstrated that few newsroom managers perceive significant obstacles to the advancement of African American journalists (and, by extension, other journalists of color). But strong majorities of African American journalists believe that such obstacles exist. Today, eight years after the publication of that study, this is still the number one issue that needs to be addressed in relation to solving the problem of minority journalist retention.

Journalists of color are not convinced that they have equal opportunities for advancement or that they are being judged by the same evaluative criteria as white journalists. Newsroom managers need to take stock as to how actions in the newsroom may potentially contribute to the development of such perceptions. Indeed, supervisors must actively strive to create a newsroom atmosphere that demonstrates that the contributions of journalists of color are equally valued and rewarded. Standard operating procedure, "fair" though it may be, is not enough-the newsroom needs to become self-conscious.

One of the findings from this research review is that supervisory relationships do not appear to be implicated among the specific factors that may be driving journalists of color away from newspaper journalism at a faster rate than whites. Rather, the issues appear to be primarily structural.

Supervisors can take heart from the fact that they don't appear to be doing anything specifically wrong that needs to be corrected. But doing nothing wrong is not the same as taking active steps to prevent dissatisfaction from developing. And supervisors are the interface between journalists of color and the potentially alienating structural features of the organization.

Interestingly, the 1997 ASNE study found that perceptions of the frequency of communications between editors and staff were the same across racial groups, but journalists of color were much more likely to cite "improving communication between management and staff, including more feedback" as a top priority for newsroom change.

This testifies to the fact that journalists of color need to see evidence that their concerns are being heard. Supervisors of journalists of color must make time to discuss with journalists of color their career aspirations, their comfort level with their job assignments and development opportunities, their concerns about the newsroom and newspaper policies regarding coverage, staffing and promotion, and whether their workload is appropriately balanced with respect to their external commitments to family and other areas of life.

The problem is rarely overt racism. Newsrooms are generally not warm and fuzzy environments, and it is unrealistic to expect the culture of the newsroom to change dramatically. But individual managers need to become more self-conscious about issues such as committee representation, story selection, story assignments, promotion opportunities, the communication of information and the career development of the people under them.

Newspapers employ few people who don't want to be newspaper journalists. Job satisfaction is generally quite high and journalists can cite various factors that motivated them to choose a career in journalism in the first place. In other words, few leave newspaper journalism because they don't like the type of work they are doing.

This strongly suggests that how the job is organized and managed plays a key role in determining whether newspaper journalists stay in or leave the field. People who leave the field tend to do so accompanied by the feeling that what they had to offer was not given a chance to bloom. Many journalists of color approach their jobs with high ideals, only to leave the field disillusioned and disappointed.

Many of the on-the-job frustrations experienced by journalists of color are also, to a great extent, experienced by white journalists. But journalists of color believe that they face unique obstacles that white journalists don't face, and in a setting that offers fewer understanding colleagues than it does for white journalists. Again, this points to the importance of the supervisory relationship in demonstrating that the news organization welcomes and appreciates the contributions of journalists of color.

Finally, journalists of color feel strongly that they have made their concerns known, but that they haven't yet been heard. Thirteen studies and 4,800 interviews over twelve years make their point. Their voices, characterized as "muted" in 1993, have evolved to "concerned" and "angry" in 2001. Journalists of color feel, with some justification, that the ball is now in the industry's court.