Skip Header

We are hiring thousands of people for the 2020 Census. Click to learn more and apply.

Working Paper Number rsm2019-01
Elizabeth Nichols, Sarah Konya, Rachel Horwitz and Andrew Raim
Component ID: #ti709510793

Abstract

In 2017, the U.S. Census Bureau conducted an experimental study to determine whether the day of the week a mailed survey invitation arrives at a housing unit affects the response rate to an online survey and whether staggering the mailings reduces the variation in daily calls to a telephone helpline. Prior research showed that there is increased response to Census Bureau online surveys the day the mail arrives at the home and on Mondays. We therefore hypothesized that mail arriving on Mondays might exponentially increase the overall response rate to an online survey. To study the “Monday” effect, we designed a split-panel test with a probability sample of 8,000 housing units from the U.S. For half of the sample, the mailings were sent primarily on Mondays to arrive late in the week while for the other half the mailings were sent primarily on Thursdays to arrive at the beginning of the next week. We found a significantly higher overall login rate for the panel whose mailings were intended to arrive at the beginning of the week compared with the panel whose mailings were intended to arrive late in the week, when controlling for the number of mailings received before Census Day, which was March 15, 2017. However, we could not pinpoint any particular day of the week that drove that increase and therefore the finding might have occurred due to a cumulative effect of all the mailings. We also found the staggered mailing (mailing out on Mondays and Thursdays) within this experiment reduced the variation in the number of calls per day to the helpline for some weeks compared to other similar studies without a staggered mailout.

X
  Is this page helpful?
Thumbs Up Image Yes    Thumbs Down Image No
X
Comments or suggestions?
No, thanks
255 characters remaining
X
Thank you for your feedback.
Comments or suggestions?
Back to Header