The Media and Government in the USA, Europe and China
Edited By Christ´l De Landtsheer, Russell Farnen and Daniel B. German
16 Does the Media Reduce Political Participation?
| 329 →
Appalachian State University, Boone, North Carolina, USA
M. Kent Jennings and Richard Niemi examine the period effect theory on continuity and change in political orientations in the US. Period effects through political socialization can change a generation or an entire society. These authors refer to wars, depression, and similar major events as affecting entire populations. We hypothesized that September 11, 2001 (9/11) would have the effect of renewing American’s interest in politics. The predicted specific effect would be a heightened consumption of news and participation in the 2002 off-presidential elections. The hypothesis was rejected because voter turnout was about 39%, pretty much in line with recent off-year presidential elections. News consumption patterns remain about the same as in pre-9/11 days, with increased attention to international affairs among past news consumer elites. Despite this low turnout, 9/11 appears to have had an effect on the Republican electorate. They increased their interest and turnout in the 2002 election, resulting in a Republican victory. In the 2004 national elections, turnout increased substantially bringing into question the reasons given for the lower 2002 turnout. A new hypothesis might be that a time lag exists between an event and its impact on turnout and/or that turnout for presidential elections is more susceptible to these types of events.
September 11, 2001 was a major disaster for the US, comparable to the shock generated by the Japanese attack on Pearl Harbor in 1941. In political socialization theory, an event of this magnitude is considered sufficient to produce a period effect. In the midst of continuity and change over time, a period effect would favor change (even dramatic) in political orientations and behavior affecting an entire population. In a longitudinal study between 1965 and 1973, Jennings and Niemi (1975, p. 329) found that the civil rights movement, Watergate scandal, and Vietnam War appeared to move the entire population, but particularly the youth generation, to exhibit a growing cynicism toward the political system. They cited strong socialization effects resulting from events such as the Great Depression, the civil rights movement, and similar highly salient occurrences.
A close examination of two significant events in American history (the 1861 to 1865 Civil War and the Great Depression) illustrates the effects of a period event. There was a major shift in the elections of 1860 (pre-Civil War), 1864, and 1868 (post-Civil War) from Democrats dominating presidential elections to the dawning of a period of Republican hegemony in American politics that lasted until the Depression election of 1932. Political participation also increased during the Civil War. In 1860, voter turnout was 4,685,561; in 1868 it was 5,122,440 (with Mississippi, Texas, and Virginia not participating and Florida’s legislature casting ← 329 | 330 → their electoral votes) (Diamond, 1976, pp. 271-273). The Depression also produced a shift in party fortunes and an increase in participation. In the 1928 election (pre-Depression), voter turnout was 36,790,364; in 1932, it rose to 39,749,382. Franklin D. Roosevelt (Democrat) was elected to four terms as President. Voter turnout rose from 48.9% in 1924 to 58.8% in 1940 (Diamond, 1976, pp. 288-289; Center for Voting and Democracy, 2003).
Prior to the 2002 election, the authors hypothesized that 9/11 would change the trend in the typically low voter turnout in previous elections and produce an increase. Often, the voter turnout was below 40% in recent off-presidential election years. The authors did not hypothesize a shift in party allegiance, although the possibility of this development was examined. The authors obtained the 2002 Final Midterm Pre-Election Poll conducted by the George Gallup organization to examine the demographics of potential turnout. The Gallup survey, conducted from October 31 to November 3, 2002, clearly predicted the Republican gains in the Senate and House. To examine the actual voter turnout and vote patterns, the 2002 American National Election Study was used.
Immediate and Long-Term 9/11 Effects
Immediately after 9/11, polls showed that Americans seemed to be re-evaluating the political system and their involvement in it. Prior to 9/11, approval ratings of Congress hovered below 50%, ranging from the low to high 40s. Following 9/11, Congressional approval ratings went as high as 84%. However, these figures quickly began slipping and by mid-year 2002, they fell into the 40s again.
The American public’s view of the media went through a somewhat similar transformation. In early September 2001 (prior to 9/11), 35% of respondents in a Pew Center Research for People and the Press survey stated that they felt media reported with accuracy. This rose to 46% in November 2001 but by July 2002, it fell to 35%. In early September, 54% felt the media was professional. This rose to 73% in November, but went down to 49% in July 2002. In early September, 43% surveyed thought the media stood up for America. This skyrocketed to 69% in November, but fell to 49% by July 2002 (Pew Center for People and the Press, 2003) as the media shifted focus by Spring and Summer of 2002 to cover scandals surrounding the Enron and WorldCom accounting debacles.
However, it appears that American news consumption habits changed in following international news in 2002 compared to 1998. Table 1 shows an increase by 5% among the “Very Closely” group, but a drop for “Somewhat Closely” of 2%, indicating a gain of only 3% in the combined categories. Moreover, most of the gains occurred among a small highly-educated segment of the population, including higher income, college graduates, and senior citizens. ← 330 | 331 →
Table 1: Following international news: 1998 and 2002 compared
|Year||Very Closely||Somewhat Closely|
The rating of the president remained high after 9/11. Prior to 9/11, President George W. Bush’s popularity was in the low 50th percentile. This was low, given the honeymoon effect which is supposed to accompany a new president’s first months in office (Bill Clinton’s ratings were also low). After 9/11, President Bush’s popularity went as high as 87%. Although Congressional and media approval/support ratings fell considerably, Bush’s approval remained close to or above 60% until March 2003.
There are several possible explanations for this trend. First, the President, as Chief of State, generally is central to the American people’s attitude toward the US political system. Perhaps 9/11 had deeply affected the US citizenry to the point where they gave a large good-will (rally-around-the-flag) support for the Presidency. Second, President Bush may have had something to do with the support level because he did go to war with Afghanistan, indicating to many Americans that he was acting against terrorism. If the President had done nothing, Neville Chamberlain style, his popularity may have dropped regardless of 9/11.
The 2002 Election
The turnout in the 2002 elections was 39 3%. While this figure is 2.9% higher than 1998, it is only 0.5% higher than in 1990 and is less than the 39.8% of turnout in 1982. Voter turnout in 2002 was not noticeably higher than the general trend for about two decades. The 2004 election may reveal that the trend of a slight increase in turnout that began in 2002 may continue. For presidential elections, the voting pattern in recent years has been around a 50% turnout. If the turnout is much higher than this figure in 2004, we may be witnessing a period effect. (This data was drawn from Federal Election Commission, http://www.fec.gov, Congressional Research Service, Election Data Service Inc., and State Election Offices, 2003.)
With regard to a shift in electoral preferences, the Republicans did gain seats in 2002. They gained five seats in the 435-member House and two in the 100-member Senate. This is atypical of an off-presidential election year in which the incumbent president’s party usually loses seats. In House seats, this type of loss occurred in 32 of 33 midterm elections between 1866 and 1994 (Campbell, 2003, p. 203). The Republican gain in 2002 is an interesting development; however, their net gains were small and we will have to await the 2004 elections to examine whether a shift is coming in terms of electoral allegiance. In the midterm elections from 1994 to 2002, 1994 is the only election which saw a party gain or loss of more than 10 seats in the House. Consequently, a small number of gains in the House and the Senate ← 331 | 332 → for the Republicans in 2002 is not out of line with most election results occurring for several decades (Campbell, 2003, p. 203).
A close examination of a right-before-the-election Gallup survey in 2002 suggests that 9/11 did affect voters both in terms of issues and voter turnout, albeit not very much, but in a distinct trend.
The 2002 Gallup Survey Results
During the period October 31 to November 3, 2002, the Gallup Organization conducted telephone surveys of 1,221 adults age 18 and over. The following analysis is based on our examination of a sub-sample of the survey provided by Gallup of 715 voters who were most likely to vote. Since this sub-sample focuses on the probable voters, it is the most reliable source of voter information on the 2002 election. Based on the survey results, the dynamics of the narrow Republican win (two Senate seats and five House seats) are easy to identify. The 2004 election may reveal the dynamics of a small shift, but the evidence is that the Republican support in 2002 was demographically similar to the support for George W. Bush in 2000 which constituted a victory in electoral votes, but a loss of popular vote. It is clear that Republican Party demographics held sway. Conservatives led liberals as being more enthusiastic by 65% to 60%. Males outdistanced females 57% to 52%. Higher incomes (except in the $50,000-$74,900 category) led lower income groups. Republicans led Democrats 64% to 51%. The South led other regions along with suburban and rural America. The only major variable running counter to a Republican win was education, high school or less leading all other categories as being more enthusiastic. In sum, this outcome reflects that the most likely non-Republican voters were liberal or moderate, with an income between $50,000 and $74,900, independent, from the east, from urban America, and with some college education. The most likely Republican voter was conservative, male, with an income between $30,000 and $49,900, from the south, rural America, and with a high school degree. ← 332 | 333 →
Table 2: Final 2002 midterm pre-election poll results – more enthusiastic demographics and issues
On issue evaluations, it is clear that voters favored Republicans (Table 3). The Republican Party was picked as better to control Congress. Particularly in Table 3, Iraq led as the most important issue with 61% reference. Terrorism as an issue also led over the economy. The latter issue was a negative for the Republicans since the economy is fairly flat in growth terms.
The 2002 National Election Study (NES) provided several interesting facts. For example, of all the voter education levels, the only one that did not provide the Republican candidates for the Senate and House of Representatives was the high school graduate (44.6% Senate and 50.0% House of Representatives). The most glaring exception to normally expected trends was the higher female vote for a Republican Congress (Table 4). This result can be tentatively interpreted as a vote for “security” on the part of the female voter; that is, the Republican Party is perceived to be better able to handle terrorism and Iraq. ← 333 | 334 →
Table 3:Final 2002 midterm pre-election poll results: party and presidential sup- port among the more enthusiastic voters
Table 4: Vote for US Senate and House of Representatives in the 2002 national election
|% Voted Republican|
|Grade School Education||71.4||61.5|
|College Education or Advanced Degree||56.5||54.5|
|Other and None||38.2||34.1|
|Source: American National Election Study (2002) N = 1,511 ← 334 | 335 →|
Table 5: Reasons for not following international news for respondents with moderate/low interest
Finally, we address the general problem of low media consumption and voter turnout in the US. Much is written about the decline of social capital in the US. The Pew Research Center for the People and the Press concludes that the reason people do not follow international affairs is their lack of background information in this area (Table 5).
Clearly, the main reason for not following international news is a matter of lack of knowledge for the largest number of respondents. The younger generation has more access to information through the Internet than older generations. Yet the Pew data show that they have lower levels of news consumption.
Previous generations at a similar stage in the life cycle followed the same pattern (Pew, 2002). Youth today tend to get their news online, but consumption of particularly network news as well as newspaper readership has declined. When push comes to shove, American citizens simply are not motivated to consume more international news. If it takes an incident as significant as 9/11 to motivate people to be more interested in worldwide politics, it is clear that short of war or other major incidents, very little can incite citizens to become better informed. However, they did take more interest during the Cold War when the prospect of war between the US and the Soviet Union seemed imminent. It is obvious that nothing since the Cold War has moved America to get more interested in international affairs. Moreover, they do not seem very interested in voting.
Our best answer as to why this is the case is the overall normalcy of politics and related events in America. People tend to be crisis-oriented and have mainly reacted to a war, civil war, economic depression, or other extraordinary major events that dramatically affected them. There is no reason to expect much change now, even though academicians might wish for more interest and involvement. Economically, Americans are near the top of the world’s PPP (purchasing power parity) per capita scale. This is a primary reason why we are a satisfied and perhaps complacent population, not prone to become actually involved in politics.
The 2002 Election Re-examined
We hypothesized prior to the 2002 election that 9/11 would positively affect voter turnout. Low voter turnout in American elections has been a hallmark of the process in recent decades. We felt that 9/11 would produce what in political socialization theory is referred to as a “period effect.” Period effects have been ← 335 | 336 → examined by using the US Civil War and the Great Depression as causal agents. Both of these events produced an increase in electoral participation and a shift in party alignment. While we did not hypothesize a party alignment shift, the possibility was considered prior to the 2002 election.
The 39% voter turnout in the 2002 midterm election was not seen as indicating an increase in participation. This figure is roughly in line with recent off-presidential-year elections. Also, the narrow 2-seat Senate gain and 5-seat House gain were not considered to indicate a party alignment shift in the electorate. While it is unusual for a party holding the White House to gain seats in a midterm election, the Republican gains were not really sufficient to declare a period of significant shift in electorate inclinations.
Explaining why 9/11 did not have a significant effect on the 2002 election participation is not easy. It seems that 9/11 affected the economy; particularly obvious is the decline in air travel and tourism. Americans appear to have been temporarily affected by 9/1l as evidenced by the dramatic increase in support of Congress. However, the underlying cynicism which began in the 1960s and 1970s has returned to dampen the spirit of participation. The general public is back to its mixed evaluation of Congress, an occurrence undoubtedly fueled by widespread knowledge of high interest group and personal contributions to political candidates and because of the media’s continuous coverage of scandals.
The 2004 election may show a higher participation rate and a shift to Republican Party support. Such a trend will probably come from the same demographics as the 2002 Republican victory: male, conservative ideology, higher income, religious, suburban/rural America, higher education, white, and southern region.
We propose that the period effect theory needs to be accompanied by one or two attendant factors. Without one or both of these additional factors, it appears that change will not occur. 1) The period events must have a direct impact on a large number of people. The US Civil War had one of the highest casualty rates per number of participants compared to other wars in history. Many sons and husbands never returned home. During the Great Depression, 25% of the work force was unemployed and many others stood at the brink of unemployment. But 9/11 did not leave homes nationwide with dead members or result in a nationwide loss of income. 2) For the event to have an effect independent of direct impact, it must somehow be communicated in a high-impact way. When the civil rights movement affected many people, it may very well have been due to the newness of television. When citizens used this new medium to see people being beaten with billy clubs in America, they reacted with shock. Today, the American people have seen so much violence on TV that they seldom react with sufficient sustained shock for it to have any real impact. They have viewed a war live, seen students shooting up a school, etc., to the point where, it seems, they have perhaps an initial reaction to a particularly big event, but it wears off in the absence of a direct personal impact. In short, ← 336 | 337 → people are so used to seeing violent events on TV that they are immune to any long-term impact. There is no new medium in the US today that has the initial emotional appeal as TV once did.
Table 6: Vote for US President (2000 and 2004)*
← 337 | 338 →
2004 Election Resnlts
In 2004, the minor trends seen in 2002 expanded into the highest voter turnout (an estimated 59.5%) since 1968 (60.8%). The “enthusiastic” demographics of 2002 brought a very small shift to a substantial increase in certain voter categories for George W. Bush. A close look at the initial results indicates that the percentage of voters voting for George W. Bush increased in almost every demographic category except religious “other” and advanced degree (Table 6).
The regular church-going, grade-school-educated segment of the population probably increased their support for Bush in 2004 compared to 2000. These voters (combined with the higher support particularly from female, Hispanic, older, Catholic, Jewish, and urban voters who typically vote Democrat) gave Bush the popular vote he needed to become a clear victor in 2004, compared to 2000. The percentage of the backbone of the Republican Party, the college-educated Protestant, white, and southern voters who voted for George W. Bush also increased.
Voter Turnout - 2004
Possibly the most interesting aspect of the 2004 national elections was the increase in overall voter turnout. The 2004 turnout in the United States more closely resembled the 1968 election than the voter turnout for presidential elections since the Richard Nixon versus Hubert Humphrey race (Mellnick and Pitzer, 2004).
Political/social specialists give various reasons why voter turnout interest in politics and in political news has dropped in recent years. Bennett, et al. (2004) account for the drop in political news consumption, which is related to a lack of voter turnout, to:
1. Fall of the Berlin Wall in 1989
2. Media “feeding frenzies,” “media circuses,” “drive-by-journalism,” and “attack journalism”
3. Over-attention to scandal
4. Negative campaigns (which “shrink the electorate and contribute to lessened interest in politics”) (Benett, et al., 2004, p. 94).
The spin-off of the above leads to lowered levels of attentiveness, knowledge, and participation (Bennett, et al., 2004, 94).
It could also be that voters simply were not interested in the issues. The top issues in the 2004 elections were foreign affairs (terrorism 19% and Iraq 19%), moral values (22%), and the economy (20%). It could be that 9/11, the war in Iraq, combined with moral issues (such as gay marriage, gay clergy, and gay adoption) plus the question of economic recovery re-invigorated the electorate. Certainly, the media’s general style of coverage apparently has not changed. Perhaps the reasons given by Bennett, et al. (2004) really do not necessarily apply to a public that ← 338 | 339 → experiences periods of relative inattention and participation in politics due to lack of issue interest.
American National Election Study (ANES), University of Michigan (2002).
Benett, S., S. Rhine, and R. Flickinger (2004). “The Things They Cared About: Change and Continuity in American’s Attention to Different News Stories, 1989-2002,” pp. 75-99 in Press/Politics, Winter.
Campbell, J. (2003). “The 2002 Midterm Election: A Typical or an Atypical Midterm?,” pp. 203-207 in PS Political Science & Politics, Vol. 36, No. 2.
Center for Voting and Democracy (2003). Presidential Election Voter Turnout, April 28. http://-www.fairvote.org/turnout/preturn.htm
“CNN Election Results” cnn.com/ELECTION/2004/pages/results/states/US/P/00/epolls.0.
Diamond, R. (ed.) (1976). Congressional Quarterly’s Guide to US Elections. Washington, DC: Congressional Quarterly, Inc.
Jennings, M. and R. Niemi (1975). “Continuity and Change in Political Orientations: A Longitudinal Study of Two Generations,” pp. 1316-1335 in The American Political Science Review, Vol. 69.
Mellnick, T. and W. Pitzer (November 7, 2004). “How America Voted,” p. 69 in The Charlotte Observer.
Pew Center for People and the Press (June 9, 2002). “Public’s News Habits Little Changed by September 11.”
Pew Center for People and the Press (March 28, 2003), “News Media’s Improved Image ShortLived.” http://www.people-press.org/reports/print.php3.ReportID=56.htm
The Gallup Organization, Inc. (November 3, 2002). “CNN/USA Today/Gallup Final Mid-term Pre-Election Poll.”
The data which provide the basis for this analysis were furnished to the authors by The Gallup Organization, Inc. The authors’ conclusions do not necessarily reflect the views of The Gallup Organization.