Appendix E (Components and Descriptions of Each Category of Response Rate Calculation [For a Sampling Frame of Housing Units])

To standardize the response rate calculations across countries, every country should group each sample element’s final disposition code into one of the following mutually exclusive and exhaustive categories:

 

A. Interviews

Component Description
Complete interviews
  • Respondent has finished the interview.
Partial interviews
  • The survey organization (in consultation with the coordinating center) may decide prior to the start of data collection to consider an interview to be a partial interview if at least some percentage (e.g., 80%) of applicable or crucial questions have been answered.
Total interviews
  • Sum of interviews.

 

B. Non-interviews—Eligible

Component Description
Refusals
  • It has been determined that there is an eligible respondent in the housing unit, but either they or someone else refuses the interview request.
Noncontacts
  • It has been determined that there is an eligible respondent in the housing unit, but the interviewer cannot gain access to the building, no one is reached at the housing unit, or the respondent is never available when the interviewer attempts an interview.
Other
  • It has been determined that there is an eligible respondent in the household (eligibility determined as of a particular date, e.g., the date that the household listing is taken), but at some time after the determination of eligibility, the respondent is unable to complete the interview due to reasons other than a refusal or is unable to be reached after repeated attempts. For example, the respondent may have died, been incarcerated or hospitalized, or left the country.
  • It has been determined that there is an eligible respondent in the household, but they do not speak any of the study language(s) or are permanently incapable of participating in the interview due to a physical or mental condition (e.g., senility, blindness, or deafness). Note: sample elements may be considered ineligible if the target population is defined such that respondents who do not speak the study language(s) or respondents who are unable to hear are explicitly excluded from the target population to which the study plans to makes inferences.
  • Any other eligible non-interview status.
Total non-interviews—Eligible
  • Sum of eligible non-interviews.
  • If the survey organization is unable to provide separate counts of each component but can provide the total number of eligible non-interviews, use the total.

 

C. Non-interviews—Unknown eligibility

Component Description
Unknown if household/occupied housing unit
  • The sample elements have not been attempted or worked (e.g., no interviewer is available in area, or replicates are introduced too late to work all sample elements).
  • Interviewer is unable to reach the housing unit due to weather or concerns about safety in a dangerous neighborhood.
  • Interviewer is unable to locate the housing unit (e.g., inaccurate or inadequate address/locating information).
Unknown if eligible respondent is in unit/no screener completed
  • It has been determined that there is an eligible housing unit, but the interviewer is unable to determine whether there is an eligible respondent in the unit. For example, a household member may refuse to complete the screener or there may be no one available to complete the screener when the interviewer visits the household. Note: these sample elements are not considered refusals, since only elements where it has been determined that there is an eligible respondent can be classified as refusals.
Other
  • Any other status for which eligibility is unknown.
Total non-interviews—Unknown eligibility
  • Sum of non-interviews of unknown eligibility.
  • If the survey organization is unable to provide separate counts of each component but can provide the total number of non-interviews of unknown eligibility, use the total.

 

D. Non-interviews—Ineligible

Component Description
Not an eligible housing unit
  • The sample elements are out-of-sample housing units or housing units that are incorrectly listed in the address frame (e.g., housing units are outside the primary sampling unit in which they are thought to be located).
  • The sample elements are non-residential units (e.g., businesses, government offices, institutions, or group quarters).
  • Housing units are vacant on the date that eligibility is determined. Note: sample elements may be considered eligible non-interviews if someone is present at the housing unit on the date that eligibility is determined, even if the household has moved and the unit is vacant when the interviewer returns.
  • Households are temporary, seasonal, or vacation residences (i.e., not the usual place of residence).
No eligible respondent
  • It has been determined that there is an eligible housing unit, but there is no eligible respondent in the unit. For example:
    • Residence with no one 18 years of age or older.
    • Respondent does not speak any of the study language(s) and the target population is explicitly defined such that respondents who do not speak the study language(s) are not considered part of the target population to which the study plans to make inferences (may also hold for physical or mental conditions, if the target population is explicitly defined to exclude persons who are blind, deaf, senile, etc.).
    • Respondent died before eligibility is determined.
    • Respondent is incarcerated or hospitalized (i.e., institutionalized) at the time that eligibility is determined, and remains institutionalized throughout the data collection period.
Other
  • Respondent is in a group/cell for which the quota has already been filled.
  • Any other ineligible non-interview status.
Total non-interviews—Ineligible
  • Sum of ineligible non-interviews.
  • If the survey organization is unable to provide separate counts of each component but can provide the total number of ineligible non-interviews, use the total.

 

Reference

pavement-enterprise