All frameworks and cheat sheets

Column chart

Unemployment rate by CPS rotation group, first vs eighth interview

Column chart of the US unemployment rate reported by first-time and eighth-time CPS households: 7.3% vs 6.8%, 9.3% vs 8.3%, 7.5% vs 6.1%.
The only difference between the two groups is how many times they had been asked before. Source: Krueger, Mas and Niu, NBER Working Paper 20396, 2014; Review of Economics and Statistics, 2017

A column chart of rotation group bias in the US Current Population Survey: the unemployment rate reported by households in their first interview against those in their eighth, in the same months, for 1976 to 1980, 2009 to 2013 and the first half of 2014. The gap grew from 0.5 to 1.4 points, which makes it one of the clearest published examples of panel conditioning. Research teams can use it to explain why repeat participants need a fresh-cohort control.

Where this appears

What's in this column chart

First interviewEighth interview
1976 to 19807.3%6.8%
2009 to 20139.3%8.3%
H1 20147.5%6.1%

Same months, groups designed to be equally representative. The gap grew from 0.5 to 1.4 points.

Use this image

Free to use in your docs, decks and posts under CC BY 4.0, as long as you credit Koji with a link to this page. To embed it, paste this into your page:

html
<a href="https://www.koji.so/visuals/panel-conditioning-cps-rotation-group-unemployment"><img src="https://www.koji.so/visuals/panel-conditioning-cps-rotation-group-unemployment/panel-conditioning-cps-rotation-group-unemployment.png" alt="Column chart of the US unemployment rate reported by first-time and eighth-time CPS households: 7.3% vs 6.8%, 9.3% vs 8.3%, 7.5% vs 6.1%." width="1200" height="793" /></a>
<p>Source: <a href="https://www.koji.so/visuals/panel-conditioning-cps-rotation-group-unemployment">Unemployment rate by CPS rotation group, first vs eighth interview</a>, Koji</p>