Status of Journalism and News Media in the Nation’s Secondary Schools
IU School of Journalism Professor
IU School of Journalism Professor
Nearly 97 percent of the United States’ high schools have at least one of the following media-related activities: a journalism course for credit, a yearbook, a newspaper, a news magazine, or a television or radio station.
This finding, among many others, is a result of a random national sample of high schools. Methods of the study are explained in the final section of this report.
With 96.9 percent of the nation’s schools reporting some type of journalistic activity, it means that about 17,230 of the country’s 17,781 country’s high schools offer communications outlets or classes for students.
More than 77 percent of the journalism educators in the survey reported that at least one journalism class was offered at the school. About 90 percent of the schools offered a media laboratory for student journalists who work on newspaper, yearbook or broadcast facilities.
Projected onto the high school population in the U.S., these percentages mean that about 453,576 students are enrolled in a course called "Journalism," and nearly half a million students serve on school media staffs.
Graph 1 shows the breakdown of schools offering different types of media.
All four news media areas have grown a bit in the past decade while literary magazines, often sponsored by English departments, have slipped (from 37.8 percent in 1991 to 34.5 percent in 1998).
Yearbook publishing remains strong, with 95.3 percent of the nation’s schools involved in what amounts to an annual magazine.
Newspapers are published in nearly 80 percent of the country’s high schools, and when combined with news magazines (5.1 percent), some type of print publication comes out at least monthly in nearly 85 percent of the nation’s schools. This compares with about 83 percent in the 1991 study.
One major growth area, as might be expected, is in radio or television news production. In part because of the development of Channel One in many high schools, which brought television technology to both classrooms and school production areas, we find that electronic media have nearly doubled since 1991 (12.7 percent in 1991 compared with 22.2 in 1998).
We note anecdotally that some schools with TV equipment have forsaken the dreaded "morning announcements" by the principal on the public address system in favor of student-produced morning news that includes video clips of interviews, games, concerts and other student – and school-related events.
Average lab size for all media operations is 30.6 students, down just a bit from 34.6 in 1991. An average-sized Journalism class is now 33 students, up a couple students from 1991.
Reduction in Journalism Programs
Despite the overall strength of news media outlets, Journalism classes and labs in schools, a number of schools that had programs in the 1990-1991 school year no longer had them in the 1997-1998 school year. This does not mean the schools reporting some discontinuance have no classes or outlets at all; it means simply that any one of a number of journalistic outlets or classes was dropped. Even with the reductions, 96.9 percent of all U.S. high schools offering at least one journalistic class or outlet.
Nearly 17 percent of the schools dropped a Journalism credit course; 12.3 percent of the schools dropped yearbook; 17.5 percent stopped publishing the newspaper; 5.4 percent discontinued a TV or radio station; and 2.1 percent quit producing a news magazine.
Given the growth in overall journalistic media from 1991 to 1998, it means that several schools not having such media or classes nearly a decade ago began offering them in the meantime.
In three major areas examined, we noted nearly the same percent of schools offering Journalistic credits.
See Graph 2 for a comparison of types of credits offered in schools in both 1998 and 1991.
Almost 66 percent of the schools in 1991 offered some type of credit for classes in Journalism; nearly 65 percent offered credits in 1998.
More than 30 percent of the schools now offer an advanced journalism course, which is up from about 25 percent in 1991.
As might be expected from the high percentage of schools that publish a yearbook, a solid number – more than 55 percent – offer credit in a yearbook class or lab. Newspaper lab or class credits are available in about 41 percent of the schools, compared with nearly 42 percent in 1991, and "other" media credits (largely radio/TV) are offered in nearly 24 percent of the schools, a solid increase from the 17 percent that offered non-print credit in 1991.
Schools accept credit in a variety of ways: 9.9 percent accept Journalism as an English requirement (compared with 13 percent in 1991); nearly 28 percent accept it as a language arts elective (up from 26 percent); 50.2 percent accept it as a general elective toward graduation (compared with 43.4 percent); 4.1 percent list "other" credit (same as in 1991) – such as vocational education, social studies, art, and the like; and a little more than 10 percent (compared with nearly 14 percent in 1991) do not grant credit for media labs.
In 1991, 89 percent of the media labs included computers for student use. In 1998, the percent was 98.5. Also, the following percentages of journalism students are using computers for these functions in 1998: e-mail, 24.5 percent; World Wide Web searches, 44.8 percent; pagination programs for design, 83 percent; and spreadsheet programs for data analysis and bookkeeping, 31.7 percent.
Recruitment of Students (Classes)One of many major concerns often expressed by journalism educators is the health of the program in terms of student participation. How are students recruited for future journalism classes? How are they subsequently recruited to be on staff?
In terms of classes, the No. 1 method employed at U.S. high schools seems to be through prospective student applications. Even though the percentage slipped several points compared with 1991, in 1998 58 percent of the journalism educators reported using this method.
Graph 3 shows methods used to recruit students into journalism classes.
While the outright recruiting by journalism educators distinguishes journalism classes from most others in the school where no direct approaches are made before registration, other methods are also employed.
For example, recommendations from English teachers are used in 35.1 percent of the schools (compared with 39.6 percent in 1991), and nearly 32 percent of the programs rely on recommendations and scheduling from guidance counselors (compared with about 31 percent in 1991).
Because journalism classes are usually electives, and because the health of those classes depends on recruitment, it would seem that effective journalism educators, like winning athletic coaches or award-winning music teachers, recruit students in a variety of ways, as seen in Graph 3.
In nearly 20 percent of the nation’s schools, journalism teachers also visit English classes for recruiting purposes.
Recruitment of Students (Publications Staff)When it comes to recruitment of publications staff members, the most common selection procedure is for the adviser to select from among applicants. More than 55 percent of the nation’s journalism teachers use the application process (down from nearly 62 percent in 1991).
Graph 4 shows other methods of recruiting.
In about 30 percent of the schools, journalism advisers select new staff members from among those who have taken classes with them or whom they know from other teachers’ classes. In more than 8 percent of the schools, new staffers are selected by a vote of the current staff. And in more than 25 percent of the schools, some other methods are used for staff recruitment.
Minority Student ParticipationGenerally, the country’s high school news media staffs are under-represented when it comes to minority student participation. In the surveyed schools, journalism educators report that 15.5 percent of the staffs are comprised of students of color. The schools they attend, however, report 22.2 percent of the student body is comprised of students of color.
Further indications of under-representation on media staffs: In this survey, 29 percent of the schools did not have any students of color on the staff or in Journalism classes. However, in those same schools, only 4.4 percent had no students of color within the general student body. Of the 71 percent of the schools having students of color in the Journalism program, 21.6 percent of the students involved were from minority backgrounds.
By contrast, a U.S. Department of Education study shows that public elementary and secondary schools included 35.2 percent minority students in 1995. An apparent reason for the rather large difference in overall school populations of minority students is that the journalism survey was randomly drawn from all U.S. secondary schools regardless of school population.
While this sample is generalizable to that group of schools, it is not generalizable to the overall population characteristics of students in attendance. For example, our sample treats large schools and small schools alike even though large schools might be more apt to have journalism programs, and large city schools are more likely to have higher percentages of students of color than are smaller rural schools.
As will be seen in the next section on characteristics and working conditions of journalism educators, however, it can be observed that the percentage of minority media educators is far below that of secondary school teachers generally and even farther lower than the percentage of minority students in the programs.
Academic BackgroundJournalism educators have earned many degrees – but only 15 percent of the total have been earned in journalism.
Of the 652 journalism educators who participated in this study, 13.2 percent held associate degrees; 95.5 percent held bachelor’s degrees; 52.4 percent held master’s degrees; and 12.3 percent held specialist (post-master’s) or doctoral degrees.
Here’s a breakdown of each degree category and the percentage of journalism majors in each among educators with a major in journalism or journalism education (percents are not cumulative): associate, 15 percent; bachelor’s 10.7 percent; master’s, 4.4 percent; and specialist’s or doctorate, 7.1 percent.
A much higher percentage of the educators in the study, 26.2 percent, indicated they held state teaching certification in journalism. This is down from 28.2 percent in 1991. Some states do not certify teachers in journalism at all; others have minimal requirements. It is conceivable that several of the 26.2 percent educators in this study who hold state certification credentials might have taken one or two journalism courses, accompanied by a teaching major in some other related field like English, and qualify for state certification in journalism.
Graph 5 includes the top four areas of certification that current journalism educators hold.
The No. 1 field of certification among journalism teachers and advisers is English (80 percent); Journalism is next with 26.2 percent. The third most frequent area of certification for those teaching journalism and advising media is social studies (20.4 percent), while speech/drama is fourth highest of the areas with 9 percent of the respondents holding credentials.
As Graph 5 shows, these areas parallel the same four areas in 1991.
Initial Interest in TeachingOne possible explanation for the relatively small incidence of formal certification among journalism educators can be observed when examining the time at which they first considered it. In answer to the question "When did you first think about getting involved in journalism education?" the largest response, 40.2 percent, was "after assignment by an administrator." In 1991, 43.2 percent got involved in this way. Thus, a plurality of journalism educators don’t get involved through their own interest or dedication in the field but through administrative decision-making or necessity.
Graph 6 shows comparisons involving current journalism educators’ initial interest in that field.
The second-most common time of first thinking about getting involved in journalism education was after the educators were in college (21.1 percent). That means more than 61 percent of all secondary school journalism teachers/advisers did not consider journalism until after college or until after being assigned to it.
Another 21 percent of the journalism educators considered journalism for college study during high school; only 16 percent considered it at any time during college; and only 2 percent considered it before high school.
By way of contrast, another item on the survey asked, "When did you first think about entering teaching as a career?"
Graph 7 shows that more than 53 percent considered teaching in general before or during high school.
Almost 28 percent of today’s journalism educators first thought of getting into teaching while in college, and only about 19 percent considered teaching after being graduated from college.
People now involved in journalism education generally come to the area after their initial interest in becoming teachers, as the two previous graphs show. While more than 81 percent of them first thought about becoming teachers before college graduations, less than 40 percent of the teachers had considered journalism education before leaving college.
Media and Non-teaching BackgroundHigh school journalism educators, despite some weaknesses in formal journalism background, seem to have other media-related experiences that might complement their current duties.
For example, 41.3 percent of the respondents claimed to have been on the staff of a publication while they were in high school. Average number of years on staff was 2.3 Other percentages of journalism educators’ media work and average years of experience include: college publications, 35.2 percent (nearly 1 year); and professional media, 27 percent (4.9 years).
More than one-third of the teachers in the study had work experience after college in areas other than teaching. In fact, these 36 percent of the journalism educators reported an average of 7.2 years of experience doing non-teaching jobs.
Years of ServiceHigh school journalism educators have been involved in media teaching or publications work for an average of nearly 9 years (compared with 8.4 years in 1991). However, they have been educators for nearly 15 years (14.6 years in 1991). This compares with an average tenure of 15.2 years for all secondary school teachers.
Graph 8 shows years as journalism educators compared with total years teaching.
The heaviest concentration of high school media educators are in their first year of doing journalistic assignments while 30.3 percent are in their first three years, and nearly 60 percent are in their first eight years of media teaching/advising. By way of comparison, only 2.6 percent of all respondents are in their first year of teaching (regardless of area taught), 11.2 are in their first three years, and only 31.4 percent are in their first eight years of total teaching experience.
Once again, these trends indicate that staffing needs cause administrators to select language arts (and other) teachers to fill in as journalism instructors and media advisers.
Also, as seen in Graph 8, as years of teaching experience increase, the lesser likelihood there is that teachers hold any journalism duties. For example, 32 percent of all teachers in this study have more than 20 years of teaching experience; however, less than 10 percent of respondents with more than 20 years experience continue to work as journalism educators.
In this study, the veteran teachers with 15 or more years experience comprised 48.2 percent of the total, but only 22.4 percent of the teachers taught journalism or advised media for 15 or more years.
Age, Gender and RaceAge. Secondary school journalism educators’ average age is 42.1 – with a median of 44. The average age of all U.S. secondary school teachers is 42.8.
Gender. Women predominate the secondary school journalism scene, much as they do in high school teaching generally.
Graph 9 shows similarities of comparisons among journalism and non-journalism educators in both 1998 and 1991.
In the current study of secondary school journalism teachers, 72.3 percent were women. The U.S. Department of Education reports that 73 percent of all secondary school teachers in public and private schools are women.
A slight increase in the number of female teachers is noted from 1991 to 1998, but the difference is not statistically significant.
Race. Teachers and advisers in journalism programs are decidedly from Caucasian backgrounds, with only 5.3 percent of the journalism educators being people of color. This is a slight increase since 1991 when only 4.3 percent of the educators represented minority groups.
Graph 10 shows the slight growth also in secondary school teachers generally.
In a recent U.S. Department of Education study, 14 percent of public and private school teachers represented minority groups as defined by the government (Black, Hispanic, Asian or Pacific Islander and American Indian or Alaskan). In 1991, 13.1 percent of the teachers were people of color.
As seen earlier, 22.2 percent of the total population of students in the journalism educators’ schools come from minority backgrounds – and 15.5 percent of the students in journalism are from these diverse backgrounds.
Clearly, teachers from diverse backgrounds are under-represented in high school journalism programs – as are minority students within the programs.
Salaries, Compensation and Family IncomeSalaries. Secondary school journalism educators seem to be lagging behind – by nearly $3,500 – in contracted school-year salary when compared with all U.S. secondary school teachers.
Journalism educators in public and private schools average $35,994 in annual pre-tax salary compared with $39,385 for all public school teachers during the 1997-1998 school year. Part of the difference is explained by the absence of private school teacher salaries included within the comparison. Traditionally, these salaries are lower than those of public school elementary and secondary school teachers, and would no doubt bring the $39,385 figure down if included. Also, this survey was of randomly selected but representative high schools across the country. Thus, a larger percentage of teachers are included from rural and small-town schools where salaries tend to be lower.
Compensation for Advising. High school journalism educators derive additional income through their media advising duties, as reflected in Graph 12 below:
Yearbook and newspaper subsidies are virtually the same at $1,302 and $1,451 respectively. While the news magazine compensation appears to be considerably lower, at $836, one must take into account that a small percentage of journalism educators advise what could be termed "news magazines."
TV and radio advisers, by contrast, seem to derive much less compensation for their duties with electronic media than do print advisers. Perhaps a high percentage of them have responsibilities for student guidance during the regular school day, and thus do not earn extracurricular income.
Literary magazine advising, though falling within the purview of the language arts department, seems to be on the same scale in 1998 as is advising of other print media. In 1991, literary magazine advising was only $600 for the school year compared with $1,400 in 1998.
Journalism educators in this study had family income (before taxes) in 1997 of $62,948. This compares with pre-tax total income of $53,212 in 1990, a growth of nearly 18.3 percent in seven years. Their annual salary average of $35,994 comprised 57.2 percent of their total family income.
Advising stipends could not be calculated because of the difference in combinations of those parts of their total income.
Teaching-Advising Schedules and WorkloadJournalism educators are busy people, but less so than in 1991 when they put in about 78 hours each week on school-related duties. In 1998, they claim to be working slightly less than 70 hours per week.
A school day for a high school journalism educator involves teaching 4.94 classes – down slightly from 5.32 in 1991. However, with many schools changing to block scheduling during the 1990s, it’s understandable that the average would go down. A typical block schedule involves four 90-minute periods (approximately) rather than a typical school day of six or seven 50- or 55-minute class periods. Among the 4.94 classes per day are 1.21 journalism sections and 1.29 classes related to media production or labs for credit.
Thus, the typical journalism educator in this study spends about 51 percent of the school day on journalism-related learning activities while the remaining 49 percent of the school day is spent on English, social studies, speech/drama or some other academic area.
Beyond these formal class and lab times totaling 24.7 hours per week (compared with 26.6 in 1991), advisers report spending 5.6 hours each week advising students in their publications or media work outside the school day. This compares with 7 hours spent in 1991.
Here’s the remainder of a typical work week: 2.9 hours of hall duty, study hall, supervising detention; 3.3 hours completing forms and administrative paperwork; 11.7 hours preparing lessons, lectures, composing tests, grading papers; 4.6 hours doing background reading in various subject areas; 0.44 hours contacting employers on students’ behalf and visiting students at worksites; 2.1 hours conducting makeup work for students; 2.7 hours per week counseling students; 3.3 hours coaching athletics; 3.4 hours directing non-athletic extracurricular activities (non-journalism); 1.8 hours participating in non-school-sponsored activities with students (such as service and church/synagogue); and 1.2 hours tutoring.
Professional Memberships and InvolvementJournalism educators were asked about their membership in professional organizations.
Graph 13 shows the percent of the representative sample (n=652) who claimed membership in various journalism- and education-related groups.
Of the 14 specific names or choices (including "other") used in the 1991 study, 12 categories showed a decrease in the percentage of journalism teachers-advisers participating. Although the decrease was fairly slight in each case, this finding, coupled with the decrease of more than 8 hours in a typical work week, seems to indicate a lesser commitment in terms of time on involvement in professional groups and in time on educational tasks. However, given the extraordinarily heavy pressures involved with teaching high school, and given a work week that’s now closely approaching 70 hours, perhaps it’s best not to make too much of such comparisons – but simply to report that they exist and should be further monitored in years ahead.
Once again in 1998, the top three organizations educators belonged to were general teacher associations not affiliated with journalism.
A state press association is the most likely group to which a journalism educator belongs (nearly 28 percent) – followed by Quill and Scroll (24 percent), international honorary for high school journalists, that has chapters in more than 12,000 secondary schools in the U.S. and more than 40 other countries.
A fairly significant number of journalism educators (more than 20 percent) belong to the Journalism Education Association.
Use of Professional Media JournalsReading. Journalism educators seem to read a wide variety of journals related to secondary school media and advising.
Graph 14 shows a decline in percentage of readers of all 11 publications that were also used in the 1991 study.
Quill & Scroll magazine had the best overall readership with 32.5 percent of the journalism educators indicating that they read it regularly. A close second was Adviser & Staff (nearly 32 percent readership) published by Jostens Yearbook Co. Next best-read publications were more than 10 percentage points less than the best-read ones.
A publication not included in 1991 but added in 1998, Adviser Update, published by the Dow Jones Newspaper Fund, showed a solid readership of more than 11 percent of all journalism educators.
In Graph 14, to get the total number of readers of each publication among journalism educators in U.S. high schools, multiply the percentage of readers by 17,230 (which is the number of schools having some type of news media outlet or journalism program). This representative sample may be projected to the entire population of schools with news media. Readership, however, should not be equated with subscriptions.
Writing. Respondents were also asked, "Have you ever written an article and had it published in one of the above publications?" More than 5 percent indicated that they had. Thus, nearly 900 of the nation’s journalism educators have written and had articles published in media-related journals.
Personal Use of News MediaHigh school journalism educators appear to be regular users of local and national news media.
They were asked, "How many days in a typical week do you use the following media?" Results follow:
Local newspaper, 5.3 days per week; local newscasts on TV, 4.9 days per week; local radio newscasts, 4.4 days per week; national radio network news, 3.4 days per week. Each of these usages went down a bit from the 1991 survey of news media use.
In addition to local media, 79.8 percent of the high school journalism educators listed general-interest magazines that they read regularly (that is, almost every issue). Among the top selections of the nearly 80 percent who read magazines regularly were Time, 39.8 percent; Newsweek, 35.3 percent; People, 13.3 percent; Reader’s Digest, 11.7 percent; U.S. News and World Report, 11.4 percent; and National Geographic, 8.4 percent.
Besides local newspaper reading, 55.4 percent of the journalism educators read at least one non-local newspaper regularly (at least once a week). Among those who do read them, the most frequently mentioned were USA Today, 66.7 percent; New York Times, 49.5 percent; and Wall Street Journal, 28.4 percent.
Advising FreedomSince Jan. 13, 1988, when the U.S. Supreme Court handed down the Hazelwood v. Kuhlmeier decision that formally allowed school officials to more easily censor school-sponsored publications, many journalism educators have been more cognizant of First Amendment issues. However, it seems that with the passage of time, fewer journalism educators think there is as much freedom allowed by administrators.
Graph 15 shows amounts of advising freedom usually afforded those who advise student publications.
In 1998, 78.6 percent reported "a great deal" or "almost complete" freedom in advising; in 1991, nearly 84 percent indicate these amounts of freedom – 5 percentage points higher than the most recent survey. Another 20.4 percent indicated "some" freedom in 1998, while in 1991 that figure was only 15 percent. Clearly, advisers have indicated a slippage in their perceived advising freedom within the decade of the 1990s.
Also, a direct question was asked journalism educators in the current study that was asked in 1991 and by the Gallup Poll organization of high school teachers (1989) and of the U.S. public (1988). The Hazelwood-related question: "The U.S. Supreme Court (in 1988) ruled in favor of more authority for high school principals to censor school-sponsored student publications. Do you believe that this was a good ruling or a bad ruling?"
Graph 16 shows the results.
In the latest survey of high school journalism educators, more than 38 percent thought Hazelwood a good ruling while only a little more than 32 percent thought so in 1991. However, both are considerably lower than the percentage of their public high school teacher colleagues (non-journalism educators), who agreed at a 71 percent rate. Interestingly, 59 percent of the general public – a much lower percentage than non-journalism public school teachers – favored the decision.
Somewhat curious is the high percentage (20.3 percent in 1998; 18.2 percent in 1991) of journalism educators in the present study who had no opinion about Hazelwood. One possible explanation could be that many advisers today were not advising more than 12 years ago when the decision was rendered. Perhaps they did not take note of it at the time or have not studied it upon taking on the role of publications or media adviser. Certainly this is a possible negative repercussion of hiring uncertified educators to either teach journalism or to advise student media.
In neither year did a majority of journalism educators think Hazelwood was a bad ruling, though more of them thought so in 1991 (49.3 percent) than in 1998 (41.5 percent). Given the large degree to which the Court favored administrative control over student media, it is surprising that more journalism educators did not disagree with the ruling in 1991. That the percentage of journalism educators disagreeing slipped nearly 8 points from 1991 until 1998 shows that fewer journalism educators favor a free, uncensored student press.
A related question was also asked: "How has the U.S. Supreme Court decision in 1988 (Hazelwood v. Kuhlmeier) affected student freedom of expression as applied to their work on official school publications in your school?"
Graph 17 shows the results.
More than 15 percent of the journalism educators in the study noted less freedom, almost 60 percent answered "no change," and almost 24 percent did not know. In 1991, 11.6 percent noted less freedom, almost 74 percent answered "no change," and nearly 14 percent did not know.
Once again, the differences between the latest survey and the one done nearly a decade earlier show that journalism educators have detected a slippage in student freedoms as applied to work on school publications.
Job SatisfactionHigh school journalism educators are well-satisfied with their jobs. More than 41 percent claim to be "very satisfied," while an additional 43.9 percent note that they are "fairly satisfied" – for a combined overall satisfaction percentage of 85.4. In 1991, 83.8 percent of the respondents reported being satisfied with their jobs in journalism education.
The latest survey shows an increase in those educators who are very satisfied with their jobs. In 1998, 41.5 percent agreed that they were "very satisfied" whereas only 35.9 percent claimed to be "very satisfied" with their journalism jobs in 1991.
These findings show higher rates of satisfaction than shown by high school teachers generally. For example, a 1987 study showed that only 70 percent of the high school teachers in the U.S. were very or fairly satisfied with their jobs, and only 9 percent were "very satisfied" – a significantly lower percentage than the journalism educators in both the 1998 and 1991 studies.
To further measure commitment to their teaching jobs, respondents were also asked this question: "Suppose you could go back to your college days and start over again; in view of your present knowledge, would you become a teacher?" Nearly 70 percent claim they would, another 18 percent say chances are about even that they would, and fewer than 13 percent say they would not.
Nearly 65 percent indicate they wish to remain in teaching until retirement eligibility, and another 18 percent say they will continue teaching unless something better comes along. Only 3 percent indicate they have plans to leave teaching as soon as they can.
MethodAll 17,781 secondary schools that included at least grades 10-12 and were listed in Patterson’s American Education 1997 were potential sources. Computers at Quill and Scroll headquarters, at The University of Iowa, include addresses for all the nation’s high schools. From these data banks, we randomly selected 1,980 schools for the study. A seven-page survey was addressed to the journalism educator. A postage-paid, self-addressed envelope was included in each, and after an initial mailing in February 1998, a follow-up to non-respondents was mailed in April 1998.
Altogether, 669 school personnel returned the survey for a response rate of almost 34 percent. By contrast, in 1991 when the same basic survey, cover letter and mailing procedures were used, we received 834 surveys for a return of nearly 44 percent. This comparison is disturbing – and perhaps is a result of a society with increasing degrees of information overload, unwelcome phone solicitors and other intrusions on people’s time.
However, we are confident in the survey’s general validity. For example, several demographic findings were similar to those gathered in 1991. Also, with 669 respondents, the maximum sampling error for a random sample of this size is 3.7 percentage points at the 95 percent confidence level. Tolerances in sampling error were smaller than that as responses moved away from the 50th percentile.