Unextrapolated Interview Summaries • tidycreel

Introduction

Creel surveys produce two types of summary products: unextrapolated summaries and extrapolated estimates. This vignette covers unextrapolated summaries — raw tabulations of interview records that describe the composition of interviewed parties without applying survey-design weighting.

When to use unextrapolated summaries:

Describing who was interviewed (angler type, method, species sought)
Checking for refusals and participation rates
Examining trip-length distributions
Computing catch-while-sought and harvest-while-sought rates as quality indicators

All unextrapolated functions are interview-weighted, not pressure-weighted. A day with many anglers and few interviews has the same weight as a day with few anglers and many interviews. For pressure-weighted, design-correct estimates, use estimate_catch_rate(), estimate_total_catch(), and related functions described in the “Interview-Based Catch Estimation” vignette.

Build the Design

Start by assembling a complete design using all v0.5.0 data layers:

library(tidycreel)

# Load example datasets
data(example_calendar)
data(example_counts)
data(example_interviews)
data(example_catch)
data(example_lengths)

# Step 1: Define the survey calendar and stratification
design <- creel_design(example_calendar, date = date, strata = day_type)

# Step 2: Attach instantaneous count observations
design <- add_counts(design, example_counts)
#> Warning in svydesign.default(ids = psu_formula, strata = strata_formula, : No
#> weights or probabilities supplied, assuming equal probability

# Step 3: Attach interview records (all extended v0.5.0 fields)
design <- add_interviews(design, example_interviews,
  catch          = catch_total,
  effort         = hours_fished,
  harvest        = catch_kept,
  trip_status    = trip_status,
  trip_duration  = trip_duration,
  angler_type    = angler_type,
  angler_method  = angler_method,
  species_sought = species_sought,
  n_anglers      = n_anglers,
  refused        = refused
)

# Step 4: Attach species-level catch data
design <- add_catch(design, example_catch,
  catch_uid     = interview_id,
  interview_uid = interview_id,
  species       = species,
  count         = count,
  catch_type    = catch_type
)

# Step 5: Attach fish length data (individual harvest + binned release)
design <- add_lengths(design, example_lengths,
  length_uid     = interview_id,
  interview_uid  = interview_id,
  species        = species,
  length         = length,
  length_type    = length_type,
  count          = count,
  release_format = "binned"
)

print(design)
#> 
#> ── Creel Survey Design ─────────────────────────────────────────────────────────
#> Type: "instantaneous"
#> Date column: date
#> Strata: day_type
#> Calendar: 14 days (2024-06-01 to 2024-06-14)
#> day_type: 2 levels
#> Counts: 14 observations
#> PSU column: date
#> Count column: effort_hours
#> Party-size term: not carried
#> Count type: "instantaneous"
#> Survey: <survey.design2> (constructed)
#> Interviews: 22 observations
#> Type: "access"
#> Catch: catch_total
#> Effort: hours_fished
#> Harvest: catch_kept
#> Trip status: 17 complete, 5 incomplete
#> Angler type: angler_type
#> Angler method: angler_method
#> Species sought: species_sought
#> Party size: n_anglers
#> Refused: refused
#> Survey: <survey.design2> (constructed)
#> Catch Data: 42 rows, 3 species
#> caught: 8, harvested: 17, released: 17
#> Length Data: 20 rows, 3 species, length: 155–512 mm
#> harvest: 14 individual
#> release: 6 binned
#> Sections: "none"

The print() output shows all five data layers: calendar, counts, interviews, catch, and lengths.

Interview Participation

Use summarize_refusals() to understand how many potential interviewees declined to participate. High refusal rates can bias estimates if refusers differ systematically from participants.

summarize_refusals(design)
#>   month participation  N percent
#> 1  June      accepted 22     100

All 22 interviews in example_interviews were accepted (no refusals). In field surveys, a refusal rate above 20% warrants investigation.

Interview Composition

Seven tabulation functions describe the composition of the interview sample.

Day Type

summarize_by_day_type(design)
#>   month day_type  N percent
#> 1  June  weekday 13    59.1
#> 2  June  weekend  9    40.9

Angler Type

summarize_by_angler_type(design)
#>   month angler_type  N percent
#> 1  June        bank 13    59.1
#> 2  June        boat  9    40.9

Fishing Method

summarize_by_method(design)
#>   month     method  N percent
#> 1  June artificial  7    31.8
#> 2  June       bait 10    45.5
#> 3  June        fly  5    22.7

Species Sought

summarize_by_species_sought(design)
#>   month species  N percent
#> 1  June    bass  6    27.3
#> 2  June panfish  5    22.7
#> 3  June walleye 11    50.0

Successful Parties

summarize_successful_parties() counts parties that caught at least one fish of their target species (as recorded in catch data), broken down by angler type and species sought.

summarize_successful_parties(design)
#>   angler_type species_sought N_total N_successful percent
#> 1        bank           bass       5            1    20.0
#> 2        bank        panfish       3            1    33.3
#> 3        bank        walleye       5            4    80.0
#> 4        boat           bass       1            1   100.0
#> 5        boat        panfish       2            0     0.0
#> 6        boat        walleye       6            3    50.0

Trip Length Distribution

summarize_by_trip_length(design)
#>    trip_length_bin N percent
#> 1            [0,1) 2     9.1
#> 2            [1,2) 5    22.7
#> 3            [2,3) 8    36.4
#> 4            [3,4) 4    18.2
#> 5            [4,5) 3    13.6
#> 6            [5,6) 0     0.0
#> 7            [6,7) 0     0.0
#> 8            [7,8) 0     0.0
#> 9            [8,9) 0     0.0
#> 10          [9,10) 0     0.0
#> 11             10+ 0     0.0

The trip-length distribution is useful for detecting outliers and understanding effort patterns.

Catch-While-Sought (CWS) and Harvest-While-Sought (HWS) Rates

CWS and HWS rates measure the proportion of interviews where anglers targeting a species actually caught (CWS) or kept (HWS) that species. These are useful quality indicators and input parameters for population models, but should not be treated as pressure-weighted catch rates for management use — they reflect the interview sample composition, not fishing pressure across the survey period.

CWS Rates by Species Sought

summarize_cws_rates(design, by = species_sought)
#>   species_sought  N n_unknown_target n_unknown_effort n_nonpositive_effort
#> 1           bass  6                0                0                    0
#> 2        panfish  5                0                0                    0
#> 3        walleye 11                0                0                    0
#>    mean_rate         se    ci_lower  ci_upper
#> 1 0.05208333 0.05208333 -0.08180114 0.1859678
#> 2 0.23333333 0.23333333 -0.41450386 0.8811705
#> 3 0.46158009 0.23247889 -0.05641517 0.9795753

CWS Rates Collapsed Across All Groupings

summarize_cws_rates(design, by = NULL)
#>    N n_unknown_target n_unknown_effort n_nonpositive_effort mean_rate        se
#> 1 22                0                0                    0 0.2980249 0.1298807
#>     ci_lower  ci_upper
#> 1 0.02792314 0.5681266

HWS Rates

summarize_hws_rates(design, by = species_sought)
#>   species_sought  N n_unknown_target n_unknown_effort n_nonpositive_effort
#> 1           bass  6                0                0                    0
#> 2        panfish  5                0                0                    0
#> 3        walleye 11                0                0                    0
#>   mean_rate        se    ci_lower  ci_upper
#> 1 0.0312500 0.0312500 -0.04908068 0.1115807
#> 2 0.1333333 0.1333333 -0.23685935 0.5035260
#> 3 0.3887987 0.1322997  0.09401668 0.6835807

Interpretation guidance: CWS and HWS rates > 1.0 are possible when anglers catch multiple fish of the target species in a single interview. A CWS rate of 0.6 means 60% of interviews targeting that species resulted in at least one catch.

Length Frequency Distributions

Length data attached via add_lengths() can be summarized by catch type and species.

Harvest Lengths

summarize_length_freq(design, type = "harvest", by = species, bin_width = 25)
#>   species length_bin N percent cumulative_percent
#> 1    bass  [225,250) 1    25.0               25.0
#> 2    bass  [250,275) 1    25.0               50.0
#> 3    bass  [275,300) 2    50.0              100.0
#> 4 panfish  [150,175) 2   100.0              100.0
#> 5 walleye  [375,400) 2    25.0               25.0
#> 6 walleye  [400,425) 3    37.5               62.5
#> 7 walleye  [475,500) 1    12.5               75.0
#> 8 walleye  [500,525) 2    25.0              100.0

Release Lengths

Release lengths in example_lengths are stored in pre-binned format. summarize_length_freq() handles this automatically:

summarize_length_freq(design, type = "release", by = species)
#>   species length_bin N percent cumulative_percent
#> 1    bass  [275,276) 4    44.4               44.4
#> 2    bass  [325,326) 5    55.6              100.0
#> 3 panfish  [175,176) 6    66.7               66.7
#> 4 panfish  [225,226) 3    33.3              100.0
#> 5 walleye  [375,376) 2    40.0               40.0
#> 6 walleye  [425,426) 3    60.0              100.0

All-Catch Lengths

summarize_length_freq(design, type = "catch", by = species, bin_width = 25)
#>    species length_bin N percent cumulative_percent
#> 1     bass  [225,250) 1     7.7                7.7
#> 2     bass  [250,275) 1     7.7               15.4
#> 3     bass  [275,300) 6    46.2               61.6
#> 4     bass  [325,350) 5    38.5              100.1
#> 5  panfish  [150,175) 2    18.2               18.2
#> 6  panfish  [175,200) 6    54.5               72.7
#> 7  panfish  [225,250) 3    27.3              100.0
#> 8  walleye  [375,400) 4    30.8               30.8
#> 9  walleye  [400,425) 3    23.1               53.9
#> 10 walleye  [425,450) 3    23.1               77.0
#> 11 walleye  [475,500) 1     7.7               84.7
#> 12 walleye  [500,525) 2    15.4              100.1

Extrapolated Species-Level Estimates

For design-correct, pressure-weighted estimates broken down by species, use the extrapolated estimators. These combine effort estimates (from count data) with species-specific catch rates (from interview + catch data) using the ratio-of-means estimator.

Catch Per Unit Effort by Species

estimate_catch_rate(design, by = species)
#> Warning: Small sample size for CPUE estimation.
#> ! Sample size is 17. Ratio estimates are more stable with n >= 30.
#> ℹ Variance estimates may be unstable with n < 30.
#> Small sample size for CPUE estimation.
#> ! Sample size is 17. Ratio estimates are more stable with n >= 30.
#> ℹ Variance estimates may be unstable with n < 30.
#> Small sample size for CPUE estimation.
#> ! Sample size is 17. Ratio estimates are more stable with n >= 30.
#> ℹ Variance estimates may be unstable with n < 30.
#> 
#> ── Creel Survey Estimates ──────────────────────────────────────────────────────
#> Method: ratio-of-means-cpue-species
#> Variance: Taylor linearization
#> Confidence level: 95%
#> Grouped by: species
#> Unit: fish/angler-hour
#> 
#> # A tibble: 3 × 6
#>   species estimate     se ci_lower ci_upper     n
#>   <chr>      <dbl>  <dbl>    <dbl>    <dbl> <int>
#> 1 bass      0.205  0.0743  0.0590     0.350    17
#> 2 panfish   0.0558 0.0311 -0.00518    0.117    17
#> 3 walleye   0.465  0.0957  0.278      0.653    17

Total Catch by Species

estimate_total_catch(design, by = species)
#> Warning: Instantaneous counts were expanded without a period length.
#> ℹ No `period_length_col` was supplied to `add_counts()`, so the estimate is the
#>   count column summed over days.
#> ! If that column holds an instantaneous angler count, the result is in
#>   angler-days, not angler-hours.
#> ℹ Supply the period each count was randomised within: `add_counts(design,
#>   counts, period_length_col = <col>)`.
#> This warning is displayed once per session.
#> 
#> ── Creel Survey Estimates ──────────────────────────────────────────────────────
#> Method: Total Catch (Effort × CPUE)
#> Variance: Taylor linearization
#> Confidence level: 95%
#> Grouped by: species
#> Effort target: sampled_days
#> 
#> # A tibble: 3 × 6
#>   species estimate    se ci_lower ci_upper     n
#>   <chr>      <dbl> <dbl>    <dbl>    <dbl> <int>
#> 1 bass        63.2  21.9    16.5     110.     17
#> 2 panfish     27.2  10.4     5.10     49.3    17
#> 3 walleye    190.   46.0    92.3     288.     17

Total Harvest by Species

estimate_total_harvest(design, by = species)
#> 
#> ── Creel Survey Estimates ──────────────────────────────────────────────────────
#> Method: Total Harvest (Effort × HPUE)
#> Variance: Taylor linearization
#> Confidence level: 95%
#> Grouped by: species
#> Effort target: sampled_days
#> 
#> # A tibble: 3 × 6
#>   species estimate    se ci_lower ci_upper     n
#>   <chr>      <dbl> <dbl>    <dbl>    <dbl> <int>
#> 1 bass        29.8  9.84     8.82     50.8    17
#> 2 panfish     18.1  6.92     3.40     32.9    17
#> 3 walleye    131.  31.4     64.4     198.     17

Release Rate and Total Releases by Species

estimate_release_rate(design, by = species)
#> Warning: Small sample size for CPUE estimation.
#> ! Sample size is 17. Ratio estimates are more stable with n >= 30.
#> ℹ Variance estimates may be unstable with n < 30.
#> Small sample size for CPUE estimation.
#> ! Sample size is 17. Ratio estimates are more stable with n >= 30.
#> ℹ Variance estimates may be unstable with n < 30.
#> Small sample size for CPUE estimation.
#> ! Sample size is 17. Ratio estimates are more stable with n >= 30.
#> ℹ Variance estimates may be unstable with n < 30.
#> 
#> ── Creel Survey Estimates ──────────────────────────────────────────────────────
#> Method: ratio-of-means-rpue
#> Variance: Taylor linearization
#> Confidence level: 95%
#> Grouped by: species
#> Unit: fish/angler-hour
#> 
#> # A tibble: 3 × 6
#>   species estimate     se ci_lower ci_upper     n
#>   <chr>      <dbl>  <dbl>    <dbl>    <dbl> <int>
#> 1 bass      0.102  0.0407  0.0225    0.182     17
#> 2 panfish   0.0186 0.0104 -0.00173   0.0389    17
#> 3 walleye   0.140  0.0385  0.0640    0.215     17
estimate_total_release(design, by = species)
#> 
#> ── Creel Survey Estimates ──────────────────────────────────────────────────────
#> Method: product-total-release
#> Variance: Taylor linearization
#> Confidence level: 95%
#> Grouped by: species
#> Effort target: sampled_days
#> 
#> # A tibble: 3 × 6
#>   species estimate    se ci_lower ci_upper     n
#>   <chr>      <dbl> <dbl>    <dbl>    <dbl> <int>
#> 1 bass       33.4  14.1      3.43     63.5    17
#> 2 panfish     9.07  3.46     1.70     16.4    17
#> 3 walleye    58.9  18.8     18.9      98.9    17

For grouped estimates combining calendar strata with species, use by = c(day_type, species). See vignette("interview-estimation") for the complete extrapolated estimation workflow.

Summary

Function	Data required	Output
`summarize_refusals()`	refused field	month × participation × N × %
`summarize_by_day_type()`	strata	month × day_type × N × %
`summarize_by_angler_type()`	angler_type	month × angler_type × N × %
`summarize_by_method()`	angler_method	month × method × N × %
`summarize_by_species_sought()`	species_sought	month × species × N × %
`summarize_successful_parties()`	catch + angler_type + species_sought	angler_type × species × success rate
`summarize_by_trip_length()`	trip_duration	bin × N × %
`summarize_cws_rates()`	catch + species_sought	CWS rate ± SE by grouping
`summarize_hws_rates()`	catch + species_sought	HWS rate ± SE by grouping
`summarize_length_freq()`	lengths	bin × N × % × cumulative %