Kuhn, P., & Mansour, H. (2013). Is Internet job search still ineffective? The Economic Journal, 124 (December), 1213-1233.
- The study’s objective was to determine whether the finding from a previous study that Internet job searches increased unemployment duration was upheld using more recent data. Specifically, the authors compared the impact of Internet job searches on unemployment duration using survey data from 2005–2008 to the estimated impact from a prior study on Internet job searches that used 1998–2001 survey data.
- The study used a nonexperimental design to compare the impacts of Internet job searches on the duration of unemployment at two different time periods. For the first time period, the authors analyzed data from the United States Current Population Survey (CPS) from 1998 to 2001. For the second time period, the authors used data from the National Longitudinal Survey of Youth (NLSY) 1997 sample, covering 2005 to 2008.
- The study found that in the earlier period (1998–2001) unemployed adults who used the Internet for any job search activities were unemployed 22 percent longer than those who did not use the Internet, whereas in the later period (2005–2008) young adults who used the Internet for job search activities were reemployed 25 percent faster than comparable workers who did not use the Internet as part of their job search strategy. Thus, the study found that the effect of using the Internet as part of a job search strategy on unemployment durations reversed in the 10 years covered by the two studies.
- The quality of causal evidence presented in this report is low because the authors did not account for other factors that might have affected unemployment duration. This means we are not confident that the estimated effects are attributable to the use of Internet search activities; other factors might have contributed.
Features of the Study
Research suggests a large increase in the use of the Internet for job search activities in the past 20 years. In this study, the definition of Internet job search was driven by the data available from the NLSY Wave 12 survey, which asked respondents about their use of the Internet to conduct job search activities. The study included two separate analyses. In the replication study, the authors reproduced the analysis from a previous study using data from the CPS Computer and Internet Use Supplements from December 1998 to August 2000, as well as monthly CPS surveys through August 2001. The study estimated the impacts of Internet use on unemployment duration using the full sample of 4,139 unemployed adults, as well as smaller sample of 669 unemployed young adults ages 23 to 29.
In the updated novel study, the authors conducted new analyses using data collected from the NLSY 1997 sample. They administered one survey during autumn 2008 and covered the years 2005 to 2008. The sample included 840 respondents who were 24 to 28 years old at the time of data collection; were employed at the time of the survey; had started that job at any time since January 1, 2007; had experienced a period of joblessness before starting that job; and reported conducting some job search activities during that period.
In both study periods, the authors compared the duration of unemployment between survey respondents who used the Internet for job search or related activities and those who did not. The authors then compared the results from the replication and novel studies to see if the effect of Internet job search activities on unemployment duration had changed during the 10 years covered by the two periods.
- Overall, the study found that the effect of Internet job searches on unemployment durations reversed in the 10 years covered by the two studies. In the earlier period, from 1998 to 2000, the study found that unemployed adults who used the Internet for job search activities were unemployed 22 percent longer than those who did not use the Internet as a job search tool. In contrast, the study found that young adults ages 24 to 28 who were unemployed at some time from 2005 to 2007 were reemployed 25 percent faster than comparable workers who did not use the Internet as part of their job search strategy.
Considerations for Interpreting the Findings
- Although the study accounted for several factors that might affect unemployment, including the age, race, gender, and education of the survey respondents, it did not account for differences in prior employment status or earnings.
- The outcomes were not measured in the same way across the two studies. In the novel analysis, the survey was administered after the period of unemployment had ended; therefore, the self-reported duration depended on the recall of the survey respondent. In the replication study, data on job search activities were collected during the period of unemployment, and follow-up surveys determined the unemployment duration.
- In addition, the study compared results from two different demographic groups. In the novel analysis, the sample comprised young adults ages 24 to 28 who had been unemployed for some period of time immediately preceding their current job. In the replication study, the sample comprised adults of any age who were unemployed at the time of the survey.
Causal Evidence Rating
The quality of causal evidence presented in this report is low because the authors did not account for other factors that might have affected the differences in unemployment duration and Internet job search activities between survey respondents in the two time periods analyzed. This means we are not confident that the estimated effects are attributable to the use of Internet search activities; other factors might have contributed.
Kuhn, P., & Skuterud, M. (2004). Internet job search and unemployment durations. American Economic Review, 94(1), 218-32.