Log inSign up
Blog results
Showing 0 of 0 results
Stay curious! You'll find something.

Why you should have confidence in our new Vox/SurveyMonkey Economic Confidence Index

Why you should have confidence in our new Vox/SurveyMonkey Economic Confidence Index

The national economy relies on business profits, GDP growth, and technological change driving new industries forward—but what drives all of those factors? Confidence. Why does the stock market fall after a shocking election? Why are businesses hesitant to invest during recessions even though the price of labor and capital is cheap? Psychological uncertainty is an important measure of economic health—and measuring it is an important task, too.

The idea of an economic confidence index isn’t anything new. Since the 1950s, the University of Michigan has published a monthly Index of Consumer Sentiment, which is considered to be a leading economic indicator by economists at the Bureau of Economic Analysis, the Federal Reserve, and elsewhere. The Conference Board publishes the Consumer Confidence Index, Bloomberg publishes the Consumer Comfort Index (which it inherited from ABC News/Money Magazine), and Rasmussen publishes a Consumer Index.

These indexes are important—they’re a way to get more detailed information about the economy aside from GDP, employment, and profits. If people are better off than they were a year ago, expect to be better off a year from now, or are just actively thinking about making big ticket purchases, those are good signs that the economy as a whole is on the upswing.

But, there’s no magic to the idea of a confidence index. The questions are public; the calculations are simple. In fact, they likely gain value if they can be replicated, because when different methodologies provide similar results the results are more reliable.

How our methodology is different

Our Vox/SurveyMonkey Economic Confidence Index draws its sample randomly from the pool of 3 million daily respondents to user-generated surveys on the SurveyMonkey platform. SurveyMonkey’s user base is incredibly diverse, and our respondents reflect that. We have Fortune 500 companies sending HR surveys to their employees, small businesses sending feedback surveys to their clients and customers, middle schools surveying their students’ parents, and soccer moms and dads sending out surveys to find out which snacks to bring before the next game. After completing their initial survey, the selected respondents see a “thank-you” webpage inviting them to take an additional survey.

We weight our data to be representative of the U.S. population according to age, race, sex, education, and geography using the Census Bureau’s American Community Survey.

We field each survey for a full seven days starting the first Monday of each month, giving us a full week’s worth of data and a large pool of respondents.

The number of respondents we reach each month will vary, because we are sampling from the stream of survey takers continuously using the SurveyMonkey platform and not relying on a panel for our respondents. Thus far, we have received close to 10,000 responses per month, a much larger sample size than any other consumer confidence survey. Because of this large sample size, we are able to calculate separate index values for various subgroups:  those who work full-time, those who are on Medicare, etc. This wouldn’t be possible without our vast scale.

Our index calculation

We use five questions to calculate an economic confidence index—the same five that the University of Michigan asks, with slight web-friendly modifications.

We use the following formula to calculate the economic confidence index:

Vox SurveyMonkey Economic Confidence Index formula

Where A, B, C, D, E are responses to the following questions, with values of 2 assigned to positive responses (Better off, Very good, Somewhat good, Continuous good times economically, and Good), 1 assigned to middle values (Same and Mixed), and 0 assigned to negative values (Worse off, Somewhat bad, Very bad, Periods of widespread unemployment or depression, and Bad).

A = Would you say that you and your family are better off or worse off financially than you were a year ago?

  • Better off
  • Same
  • Worse off

B = Now looking ahead – do you think that a year from now you and your family will be better off financially, worse off financially, or just about the same as now?

  • Better off
  • Same
  • Worse off

C = Now turning to business conditions in the country as a whole – do you think that during the next 12 months we’ll have good or bad times financially?

  • Very good
  • Somewhat good
  • Mixed
  • Somewhat bad
  • Very bad

D = Looking ahead, which would you say is more likely to take place in the next five years for the country as a whole:

  • Continuous good times economically
  • Periods of widespread unemployment or depression

E = Thinking about the big things people buy for their homes – such as furniture, a stove, a television… Generally speaking, do you think now is a good or bad time for people to buy major household items?

  • Good
  • Mixed
  • Bad

We impute any missing index values using a sequential recursive partitioning function based on age, gender, education, income, and party identification. These imputed values are only used for index calculation, not in any crosstabs or in the full banners.

The University of Michigan index formula is similar in that it gives each question equal weight. Because their survey has undergone some methodological changes since it began in the 1950s, they correct for this with a different denominator and an additional constant added. When we plug our data into their index formula, we get different values each month, but a month-to-month percent change that exactly matches theirs.

As noted above, our goal in replicating their index is to verify their results and put our own SurveyMonkey spin on them. This index demonstrates our confidence in our data and methodology—and hopefully confidence in the economy, too.