Response Rates and Attrition
Once in the ALP, participants tend to remain indefinitely, resulting in relatively low attrition rates. To calculate attrition rates, we use the number of persons participating at least once in the household information survey in:
- a given base period and
- a later comparison period
Mathematically, the attrition rate is the difference in (a) and (b) divided by (a), or (a-b)/a. ALP attrition rates by year since 2006 are in the last column of the table below. The samples are taken to be close to January 1st of each year.
|Base year||Comparison year||Active in base period||Active in comparison period||% active||% attrition|
This table shows attrition for 2006 in 2007 was 15 percent. That is, of the 341 respondents who completed the household-information survey at least once in 2006, 51 (or 341 less 290), or 15 percent, did not do so in 2007. Similarly, it shows an attrition rate of 11 percent for 2007 respondents in 2008. The attrition rates have been between 6 and 13 percent annually.
Panel members do not always give formal notification about their intent to leave the panel. Rather, they simply stop participating in surveys over a prolonged period of time. To avoid retention of disinterested panel members, at the beginning of each year RAND attempts to contact members who were not active for more than a year to ask whether they are still interested in participating. The ALP removes all those for whom such contact attempts fail, as well as all those who were not active in the previous year.
Panel Response Rates
The removal of panel members who have been inactive for at least a year is not an issue for calculating attrition rates (because such members are not considered to be part of the active group for a base period). However, because inactive panel members are only removed once per year, the set of panel members who are invited to respond to a survey will typically include a number of panel members who should be considered inactive. In computing response rates for an individual survey, these should not be considered part of the panel anymore in order to avoid overestimation of the size of the survey sample.
To take such 'hidden' attrition into account, response rates for surveys conducted within the American Life Panel are calculated based on the following statistics:
- Size of selected sample: Samples are drawn based on the selection criteria applicable to the survey, i.e. in relationship to the goal of the research. Only ALP panel members considered active at the time of the survey are selected. A member is considered to be active if s/he completed a non-household information survey within one year prior to the fielding date of the survey. For example, if the survey was fielded on April 15, 2009, then a member is considered to be active if s/he responded to a non-household information survey between April 15, 2008, and April 14, 2009. If s/he did not respond to a non-household information survey in that period, s/he is considered to be inactive at the time of the survey and is not part of the selected sample.
- Number of completed interviews: the number of ALP panel members who completed the survey.
- Started but did not complete the survey: the number of ALP panel members who started the survey, but did not complete it.
- Did not start survey: the number of ALP panel members who were selected but did not start the survey.
We calculate the response rate for a given survey by dividing the number of completed interviews by the size of the selected sample for it.
Most persons completing the interview respond within one week of the date the survey went into the field. Almost all do so within two weeks. Response rates for an ALP survey typically average around 70 percent, but can vary significantly higher by subgroup, or how long the survey is kept in the field, and the number of reminders sent.
Sometimes members of a household who are not the head of household form their own household. In such cases they are assigned a new identifier. For more information, please contact us.