2011-01-04_PR_Packet - Laserfiche WebLink

Home

Roseville Parks and Re(:reation Task Based on NRC's experience and that of other leading researchers, we have created a table that compares the strengths of various survey administration modes. A previous version of this table was included in our book Citizen Surveys; How to do them, how to use them, what they mean (1CMA, 2000). Table 1: Comparison of administration methods for general population surveys Issue Phone Mail Web' Moderately Moderately Moderately Expense per(1-onipleted survey a, inexpensive inexpensive a, a e e si e Speed of adniinistration Moderately Moderately Moderately st slow st response rate F'air Good Poor Ability,to obtain : dii d responses [" ii E le l l e o- Ex(lellent Oiniination of interviewer bias ii r E le l l e o- Ex(lellent Ability,to get at ii - e a li o- ii"' d F' ii r Use of visual aids Poor Good Good Enfor(:enient of question order Ex(:ellent Poor Good Oiniination of"(:overage" bias F'air/Good Ex( F'air/Good im-11usion of flUterate respondents Good Poor Poor im-11usion of young adults Poor Poor F'air im-11usion of respondents of lower so(-J e: : ii : a at a e:ci f li n gs e r li k: lo(:ations Poor E :e l l e o- Poor (K- NatiuC ial Re s e&Irdh C&C it&rl 2010 For the survey project that the RPR wishes to conduct, we would explore the best fit, given the preferred cost allocation, the timeframe and the type of data. All things being equal, we often will recommend a mail survey because of the lower cost, better respondent candor, higher response rates and the ability to target respondent geographic location with the precision required to understand differences by subareas. We will give careful and impartial guidance regarding the options and will steer you away from any approach that may compromise the quality of the data. ' As a strand-alone data collection method with recruitment via mailed postcard,the response rate for Web is surprisingly low. The large mailing size required to achieve the desired number of completes significantly increases the cost of data collection via Web.Web is more economical when paired with a mailed survey(permitting responses via either method)or when a complete list of email addresses is available for the population of interest,as often is the case for members of organizations. 2"Coverage"bias refers to the bias that may be introduced into survey results based on the parts of the population that may not be reachable by a certain data collection mode. For a mailed survey,coverage is quite high. We use lists based on LISPS addresses, so nearly all housing units are included, even those in multi-family buildings. Land-line phone coverage is rapidly declining,with differential rates by socio-economic and demographic characteristics. Some of this bias can be eliminated by including cell phones,but there are issues with calling cell phones. Due to federal regulations, special dialing procedures have to be used. (Cell phone numbers cannot be computer-dialed,but must be hand-dialed.)You might want to consider whether recipients will be annoyed to hear from the City on their cell phone, even for a survey. It is much more expensive to include cell phone users,but may be necessary to get a representative sample of the community. Page 6