AdjWgt: Adjust design weights for nonsampling error.

In LAStarcevich/TrendNPS: NPS Trend Analyses for Complex Survey Designs

Description

Using the evalstatus field in the design file, adjust the design weights for frame error and nonsampling error.

Usage

AdjWgt(dat, popn, evalstatus = "EvalStatus", wgt = "wgt")

Arguments

dat	Data frame containing columns at least for Site, WYear, Year, the outcome of interest, and the survey evaluation status EvalStatus. See section "Options for variable EvalStatus" below.
popn	Takes the value of "Target" or "Sampled" to reflect inference to either the original target population or to the subpopulation of sites that would not be subject to nonresponse, respectively.
evalstatus	Status of the survey at each site relative to frame error and nonresponse error. Note that this field is calculated separately for each year at a site.
wgt	Original design weight. The design weights for a given year should sum to the size of the population before accounting for any frame error.

Details

Adjusts the original sampling design weights for frame and nonresponse error. See (Starcevich et al. 2016) for more information on weighting adjustments and populations of inference. See (Starcevich et al. 2017) for an example on applying the AdjWgt_FrameNR function to a monitoring data set.

Value

Returns a vector of adjusted design weights that should sum to the extent of the population of interest after accounting for frame error (popn = "Target") or frame and nonresponse error (popn = "Sampled").

Options for variable evalstatus

The evalstatus field requires specific formatting for weighting adjustment. Valid options include:

"Target - Surveyed"	Site is in target population and was successfully surveyed.
"Target - Not Surveyed"	Site is in target population but was not successfully surveyed.
"Non-Target"	Site is not in target population, not surveyed.
"Not Evaluated"	Site was not evaluated as target or not, not surveyed.

Author(s)

Leigh Ann Starcevich of Western EcoSystems Technology, Inc.

References

Starcevich, L. A., G. DiDonato, T. McDonald, and J. Mitchell. (2016). A GRTS user's manual for the SDrawNPS package: A graphical user interface for generalized random tessellation stratified (GRTS) sampling and estimation. Natural Resource Report NPS/PWRO/NRR-2016/1233. National Park Service, Fort Collins, Colorado.

Starcevich, L. A., T. McDonald, A. Chung-MacCoubrey, A. Heard, and J. Nesmith. (2017). Trend Estimation for Complex Survey Designs. Natural Resource Report NPS/xxxx/NRR-2017/xxxx. National Park Service, Fort Collins, Colorado.

Examples

## Not run: 
#  ---- Read example data set.
	LakeData$AdjWgt[LakeData$WYear == 0] = AdjWgt(dat=LakeData[LakeData$WYear == 0,],
	                                              popn="Target",
	                                              evalstatus = "EvalStatus",
	                                              wgt = "wgt")

## End(Not run)

LAStarcevich/TrendNPS documentation built on May 21, 2019, 9:19 a.m.