Single-phase Mail Survey Design for Rare Population Subgroups

Field Methods ◽  
2016 ◽  
Vol 28 (4) ◽  
pp. 381-395 ◽  
Author(s):  
J. Michael Brick ◽  
William R. Andrews ◽  
Nancy A. Mathiowetz
2019 ◽  
pp. 107755871988840 ◽  
Author(s):  
Q Burkhart ◽  
Nate Orr ◽  
Julie A. Brown ◽  
Ron D. Hays ◽  
Paul D. Cleary ◽  
...  

We assess the association between survey layout and response rates (RRs) in the 2017 Medicare Advantage Consumer Assessment of Healthcare Providers and Systems mail survey. Among 438 Medicare Advantage plans surveyed by six vendors, there was latitude in survey layout, and plans could add up to 12 supplemental items. Regression models predicted survey response from survey characteristics (page count, number of supplemental items, and survey attractiveness), and beneficiary sociodemographics. Beneficiary-age-by-survey-characteristic interactions assessed whether survey characteristics were more strongly related to RRs among older beneficiaries. We found that surveys with more supplemental items and less attractive layouts had lower adjusted odds of response. RRs were more sensitive to format among older beneficiaries. The difference in adjusted RRs for the most favorable versus the least favorable survey design was 14.5%. For a 65-year-old, this difference was 13.6%; for an 80-year-old, it was 21.0%. These findings suggest that even within a relatively standardized survey, formatting can substantially influence RRs.


2002 ◽  
Vol 19 (3) ◽  
pp. 303-328 ◽  
Author(s):  
James G. Helgeson ◽  
Kevin E. Voss ◽  
Willbann D. Terpening

2015 ◽  
Vol 44 (16) ◽  
pp. 3403-3427 ◽  
Author(s):  
Lee Dykes ◽  
Sarjinder Singh ◽  
Stephen A. sedory ◽  
Vincent Louis

Field Methods ◽  
2018 ◽  
Vol 31 (1) ◽  
pp. 39-55 ◽  
Author(s):  
Jennifer Dykema ◽  
Nadia Assad ◽  
Griselle Sanchez-Diettert ◽  
Kelly Elver ◽  
John Stevenson

Best practices in mail survey design advise personalizing correspondence, but most research predates the use of address-based sampling (ABS) in which a householder’s name either cannot be matched to an address or may be matched incorrectly. Further, recent evidence casts doubt on the effectiveness of personalization. The current study examines the impact of using a personalized versus a generic salutation on response rates and data quality in an ABS mail survey of the general population. A sample of 2,000 household addresses in Wisconsin was randomly selected from the U.S. Postal Service Delivery Sequence File. For households in the personalized salutation group, all materials used the surname of the household; cases in the generic salutation group referred to the state conducting the survey. While personalization was not related to response rates, it was associated with higher levels of item nonresponse. Findings support current recommendations not to use names in ABS mail surveys.


1993 ◽  
Vol 22 (9) ◽  
pp. 63-69 ◽  
Author(s):  
Ravindra Singh ◽  
N.S. Mangat ◽  
Sarjinder Singh

<i>Abstract</i>.—Fisheries managers are looking for valid information on basic characteristics of recreational fisheries, such as landings data, to inform management decisions. We present a complementary survey approach designed to generate data on effort and harvest as well as various human dimensions of anglers using a telephonediary- mail survey design for a multispecies, multi-site fishery in a water-rich state in northern Germany (Mecklenburg-Vorpommern). First, a nationwide telephone screening was applied to estimate the total number of active resident and nonresident anglers in the study area. Afterwards, a 1-year diary study with randomly recruited resident and nonresident anglers was conducted. Routine check-up telephone calls were used to encourage the participants, generate detailed human dimensions data on the characteristics of anglers, and evaluate diary entries. After the end of the diary study, 648 anglers (58%) returned complete diaries. Responding diarists were significantly older, had a higher level of education, and encompassed more avid anglers than the nonresponding participants. Thus, diarists were weighted against external characteristics of a random sample of the resident angler population to reduce the risk of biased catch and harvest estimates. Indeed, estimates for harvest and effort based on weighted samples were significantly lower than unweighted mean estimates. Extrapolations of average annual harvest rates per angler to the population level revealed that for the most economically important fish species such as European eel <i>Anguilla anguilla</i>, Atlantic cod <i>Gadus morhua</i>, northern pike <i>Esox lucius</i>, common carp <i>Cyprinus carpio</i>, or Eurasian perch <i>Perca fluviatilis</i>, recreational fishing landings greatly exceeded commercial fisheries landings. Because diary estimates of annual angler landings were generally smaller relative to estimates of angler harvest stemming from 3-month recall periods using telephone surveys and on-site creel surveys, we concluded that the use of diary data likely resulted in conservative estimates of total landings. Our survey design may serve as a model for further studies because of its cost-effectiveness relative to standard creel surveys and because the panel structure of diary studies allows rich insights into individual angler behavior that is not possible to be accomplished by cross-sectional creel surveys.


Curationis ◽  
2013 ◽  
Vol 36 (1) ◽  
Author(s):  
Leana R. Uys ◽  
Robin P. Newhouse ◽  
Arwa Oweis ◽  
Xiaokun Liang

Background: Global research productivity depends on the presence of contextual factors, such as a doctorally prepared faculty, graduate programmes, publication options, that enablethe conduct and publication of studies to generate knowledge to inform nursing practice.Objectives: The current study aimed to develop and test an instrument that measures the level of contextual support for nursing research within a specific country, allowing comparisons between countries.Method: After development of a 20-item survey with seven factors and 11 criteria based on aliterature review, a quantitative descriptive e-mail survey design was used.Results: Nurse researchers (N = 100) from 22 countries were invited to participate. Theresponse rate was 39% from 15 countries. Ethics approval was obtained by investigators in their country of origin. Results showed wide variation in the level of contextual support. The average total level of support across all countries was 26.8% (standard deviation [SD] = 14.97). The greatest variability was in the area of availability of publishing opportunities (ranging between no suitable journals in a country to over 100). The least variability was in the area of availability of local enabling support (SD = 7.22). This research showed wide differences in the level of contextual support for nursing research.Conclusion: The survey instrument can be utilised as a country assessment that can be used tostrategically plan the building of infrastructure needed to support nursing research. Contextual support for nursing research is an antecedent of strong science. Building infrastructure for nursing science is a priority for global health.


Sign in / Sign up

Export Citation Format

Share Document