Skip to contents

Construct a hybrid access + roving survey design

Source: R/hybrid-design.R
as_hybrid_svydesign.Rd

[Experimental]

Usage

as_hybrid_svydesign(
  access_data,
  roving_data,
  date_col = "date",
  strata_col = "day_type",
  count_col = "count",
  access_fraction = NULL,
  roving_fraction = NULL,
  fpc = TRUE
)

Arguments

access_data

Data frame of access-point count observations. Must contain the columns named by date_col, strata_col, and count_col.

roving_data

Data frame of roving-route count observations. Must contain the same columns as access_data.

date_col

Character scalar. Name of the date column (shared by both tables). Default "date".

strata_col

Character scalar. Name of the stratum column (shared by both tables). Default "day_type".

count_col

Character scalar. Name of the count column (shared by both tables). Default "count".

access_fraction

Named numeric vector. Sampling fraction per stratum for the access-point component (proportion of access points sampled on each sampled day; must be in (0, 1]). Names must match stratum values in access_data.

roving_fraction

Named numeric vector. Sampling fraction per stratum for the roving-route component. Names must match stratum values in roving_data.

fpc

Logical. Apply finite-population correction? Default TRUE.

Value

A survey::svydesign object with an additional class attribute "creel_hybrid_svydesign". The design data contains a component column ("access" or "roving") and a weight column derived from the sampling fractions.

Details

Combines count data from access-point and roving survey components into a single survey::svydesign object suitable for effort estimation via estimate_effort().

A hybrid design is appropriate when a survey uses both fixed access-point counts (e.g., boat-launch interviews with angler counts) and roving-route counts (e.g., progressive counts along a shoreline transect) within the same sampling frame. The two components are stacked, with a component column distinguishing them, and inclusion probabilities are derived from the user-supplied sampling fractions per stratum.

PSU alignment requirement: Both access_data and roving_data must share the same date and stratum columns. Mismatched column names, or dates present in one component but absent in the other, trigger an error rather than a silent expansion. If PSU boundaries differ (e.g., access routes cover different sections than roving routes), the estimates will be biased; a warning is issued when stratum-date combinations are asymmetric.

See also

Other "Survey Design": add_catch(), add_counts(), add_interviews(), add_lengths(), add_sections(), as_survey_design(), compute_angler_effort(), compute_effort(), creel_design(), creel_schema(), est_effort_camera(), prep_counts_boat_party(), prep_counts_daily_effort(), prep_interview_catch(), prep_interviews_trips(), validate_creel_schema()

Examples

if (FALSE) { # \dontrun{
access <- data.frame(
  date     = as.Date(c("2024-06-01", "2024-06-02")),
  day_type = c("weekday", "weekend"),
  count    = c(12L, 30L)
)
roving <- data.frame(
  date     = as.Date(c("2024-06-01", "2024-06-02")),
  day_type = c("weekday", "weekend"),
  count    = c(8L, 22L)
)
design <- as_hybrid_svydesign(
  access_data      = access,
  roving_data      = roving,
  access_fraction  = c(weekday = 0.5, weekend = 0.5),
  roving_fraction  = c(weekday = 0.4, weekend = 0.4)
)
survey::svytotal(~count, design)
} # }