skip to main content
Americans' Identification as Middle Class Edges Back Up

Americans' Identification as Middle Class Edges Back Up

Chart: data points are described in article

Story Highlights

  • 58% identify as upper-middle or middle class in 2016
  • This is up from 50% in 2012 and 51% in 2015
  • The rise in middle-class identification has occurred across income groups

PRINCETON, N.J. -- The percentage of Americans who say they belong in the upper-middle and middle class has edged up to an average 58% this year, compared with 50% in 2012 and 51% in 2015.

Trend: Americans' Social Class Self-Identification

The 2016 figure is based on the average of three separate polls conducted in September, October and December. Despite this year's recovery in middle-class identification compared with 2015 and 2012, middle-class identification remains slightly below the 61% average measured between 2000 and 2008.

This year's uptick in middle-class identification was not directly related to the presidential election, even as economic confidence rose sharply after Nov. 8. Two of the three recent surveys showing higher middle-class identification were conducted before the election, and there was little change in the third, post-election survey conducted Dec. 7-11.

Gallup's class identification question gives Americans five categories from which to choose. Across nine surveys conducted since June 2008, only between 1% and 3% of Americans say they are upper class, with the rest spread out across upper-middle-, middle-, working- and lower-class categories. In 2012 and 2015, almost twice as many Americans put themselves in the "lower class" category as did in 2008 and 2016. This, combined with a slight increase in working-class identification in 2012 and 2015, resulted in the drop in middle-class identification during those years.

Americans' Social Class Self-Identification
  2008 2012-2015 2016
  % % %
Upper class 2 2 3
Upper-middle class 18 12 15
Middle class 44 39 43
Working class 28 33 30
Lower class 8 15 8
Figures represent averages for June 2008 and two September 2008 surveys, one 2012 and one 2015 survey, and three 2016 surveys

Americans' self-identified social class does not necessarily reflect their actual socioeconomic status because in responding to the question, Americans can choose to place themselves in any category they want. However, there is a strong relationship between respondents' self-reported income and their self-placement on the social class spectrum. Americans are two to three times more likely to identify as upper-middle or middle class if they report a household income of at least $75,000 than they are if their income is under $30,000, based on aggregated data across the three time periods.

Identification as Upper-Middle/Middle Class, by Income
  2008 2012-2015 2016
  % % %
<$30,000 28 25 36
$30,000 to <$75,000 59 48 53
$75,000+ 86 78 82
National average 62 51 58
Figures represent averages for June 2008 and two September 2008 surveys, one 2012 and one 2015 survey, and three 2016 surveys

The shifts in class identification between 2008 and the present occurred across income categories. Although the patterns fluctuate, middle-class identification dipped in each income group in 2012-2015 and recovered in each group at least somewhat this year.

The shifting percentages over the past eight years in those identifying as middle class occurred mostly among those without a college degree. Middle-class identification among those with college degrees has been more stable -- and high.

Identification as Upper-Middle/Middle Class, by Education
  2008 2012-2015 2016
  % % %
Less than high school 36 29 36
High school graduate 50 38 49
Some college 60 45 53
College graduate 80 76 79
National average 62 51 58
Figures represent averages for June 2008 and two September 2008 surveys, one 2012 and one 2015 survey, and three 2016 surveys

Even though the political circumstances have changed significantly since 2008, the dip in 2012 and 2015 (compared with 2008) and then the recovery this year was fairly similar among Republicans, independents and Democrats.


With the exceptions of 2012 and 2015, Americans' subjective class identification has been fairly consistent over the past 16 years. Relatively few see themselves as lower class, and even fewer identify as upper class. That leaves the bulk of the population viewing themselves as working or middle class, with the rest in the upper-middle-class category.

Gallup did not measure subjective class identification between 2008 and 2012, and it's possible that the drop in middle-class identification in the latter year, along with 2015, reflected the lingering aftermath of the Great Recession and mortgage crisis.

The uptick this fall, across three separate surveys, could reflect many factors. The rise was not confined to just one partisan group and was not a direct result of the election of Donald Trump. Perhaps most importantly, middle-class household incomes rose last year, and the shift from lower- and working-class to middle-class identification could reflect these real-world circumstances.

Researchers over the years have made many attempts to classify who is and who is not middle class, based on income and household composition. The Pew Research Center, for example, analyzed government data and reported that the percentage objectively in the middle class has declined significantly from 1971 to last year. It is, however, still too early for analysts to determine whether the percentage of Americans who can be grouped in the middle class based on their incomes and household compositions did in fact rise in the final months of this year.

Historical data are available in Gallup Analytics.

Survey Methods

Results for the most recent Gallup polls reported in this article are based on telephone interviews conducted in surveys conducted Sept. 14-18, Oct. 5-9 and Dec. 7-11, 2016, with random samples of 1,033, 1,017 and 1,028 adults, respectively, aged 18 and older, living in all 50 U.S. states and the District of Columbia. For results based on each of these samples of national adults, the margin of sampling error is ±4 percentage points at the 95% confidence level. All reported margins of sampling error include computed design effects for weighting.

Each sample of national adults includes a minimum quota of 60% cellphone respondents and 40% landline respondents, with additional minimum quotas by time zone within region. Landline and cellular telephone numbers are selected using random-digit-dial methods.

View survey methodology, complete question responses and trends.

Learn more about how the Gallup Poll Social Series works.

Gallup World Headquarters, 901 F Street, Washington, D.C., 20001, U.S.A
+1 202.715.3030