Residential Telephone Service Survey (RTSS)

Summary of changes

Activity on this program started: 1996

Reference period of change - December 2013

The RTSS was not conducted in 2011 or 2012.

Collection instrument - For 2013, small changes were made to the question about having active cellular phones and to the question about other types of telephone service used by household. The modifications made to the questionnaire were tested informally.

Reference period of change - December 2010

The RTSS was not conducted in 2009.

For 2010, the questionnaire was modified to more accurately measure different uses of land-based phone connections, and to better distinguish between phones used for household use and those used for business purposes.

Reference period of change - December 2006

The questionnaire was modified to reflect the changing telephone service market. The questionnaire includes information on the various types of phone service provided (land-line, cellular, cable and Voice over IP (VoIP)), and the number of personal and business telephone numbers associated with the household. In 2006 only, respondents were also asked about the reasons for not having land-line phone service and the use of a cellular phone during periods without land-line telephone service.

Imputation - Usually, this methodology does not apply to this survey.

During processing of the December 2006 RTSS, it was discovered that there was an error in the way question RTS_Q12 was implemented in the field. This resulted in 17,178 respondents providing invalid responses for their total household income. These responses were automatically imputed from valid responses collected for the survey using a logistic regression technique developed in SAS to provide a more reliable estimate of total household income levels.

Overall, 43.9% of responses to RTS_Q12 are imputed. When the model was tested using valid survey responses it accurately estimated the correct value of RTS_Q12 for 84.7% of records. The model misclassified 8.7% of records as having a value of 2 (household income above the LICO) and 6.7% of records as having a value of 1 (household income below the LICO).

Reference period of change - December 2005

As of 2005, the Residential Telephone Service Survey is an annual survey.

Reference period of change - May 2003

In May 2003, the follow-up confirmation of having cell phone(s) only was introduced.

Reference period of change - May 2002

Instrument design - The questionnaire has remained the same since the start of the survey, except for the addition of the cell phone only question in May 2002.