Resources

A fisheries biologist conducts a mark-recapture survey on an isolated population of brook trout (Salvelinus fontinalis) inhabiting a subalpine tarn. The biologist marks and releases an initial sample of M=30M = 30 trout. In a subsequent sampling pass, C=25C = 25 trout are captured, of which only R=2R = 2 are marked. Because the number of recaptures is small (R<7R < 7), the standard Lincoln-Petersen estimator N^=M⋅CR\hat{N} = \frac{M \cdot C}{R} exhibits significant positive small-sample bias. To correct for this boundary condition, Chapman's modified estimator is applied: N^C=(M+1)(C+1)R+1−1\hat{N}_C = \frac{(M + 1)(C + 1)}{R + 1} - 1, with estimated variance Var⁡(N^C)=(M+1)(C+1)(M−R)(C−R)(R+1)2(R+2)\operatorname{Var}(\hat{N}_C) = \frac{(M + 1)(C + 1)(M - R)(C - R)}{(R + 1)^2 (R + 2)}. Calculate: (a) the standard Lincoln-Petersen estimate N^\hat{N}, (b) the Chapman-corrected estimate N^C\hat{N}_C, and (c) the standard error SE⁡(N^C)=Var⁡(N^C)\operatorname{SE}(\hat{N}_C) = \sqrt{\operatorname{Var}(\hat{N}_C)}.

Answer and steps

Final answer

(a) N^=375 trout\hat{N} = 375\text{ trout}; (b) N^C=268 trout\hat{N}_C = 268\text{ trout}; (c) SE⁡(N^C)=119 trout\operatorname{SE}(\hat{N}_C) = 119\text{ trout}

Steps

01

Identify the Given Survey Parameters

From the mark-recapture survey data:

  • Number of trout marked and released initially: M=30M = 30
  • Total number of trout captured in the second sample: C=25C = 25
  • Number of marked trout recaptured in the second sample: R=2R = 2

02

(a) Calculate the Lincoln-Petersen Estimate (N^\hat{N})

The standard Lincoln-Petersen estimator assumes that the proportion of marked individuals in the second sample equals the proportion of marked individuals in the entire population: RC=MN  ⟹  N^=M⋅CR\frac{R}{C} = \frac{M}{N} \implies \hat{N} = \frac{M \cdot C}{R}

Substitute the values: N^=30⋅252=7502=375 trout\hat{N} = \frac{30 \cdot 25}{2} = \frac{750}{2} = 375\text{ trout}


03

(b) Calculate Chapman's Modified Estimate (N^C\hat{N}_C)

When recaptures are small (R<7R < 7), the Lincoln-Petersen formula systematically overestimates population size. Chapman's estimator reduces this small-sample bias by adjusting the numerator and denominator: N^C=(M+1)(C+1)R+1−1\hat{N}_C = \frac{(M + 1)(C + 1)}{R + 1} - 1

Substitute the values: N^C=(30+1)(25+1)2+1−1=31⋅263−1\hat{N}_C = \frac{(30 + 1)(25 + 1)}{2 + 1} - 1 = \frac{31 \cdot 26}{3} - 1

Compute the numerator and evaluate: 8063−1≈268.67−1=267.67\frac{806}{3} - 1 \approx 268.67 - 1 = 267.67

Rounding to the nearest whole fish gives: N^C=268 trout\hat{N}_C = 268\text{ trout}


04

(c) Calculate the Standard Error (SE⁡(N^C)\operatorname{SE}(\hat{N}_C))

The estimated variance of Chapman's estimator is given by: Var⁡(N^C)=(M+1)(C+1)(M−R)(C−R)(R+1)2(R+2)\operatorname{Var}(\hat{N}_C) = \frac{(M + 1)(C + 1)(M - R)(C - R)}{(R + 1)^2 (R + 2)}

Evaluate each component:

  • (M+1)(C+1)=31⋅26=806(M + 1)(C + 1) = 31 \cdot 26 = 806
  • (M−R)=30−2=28(M - R) = 30 - 2 = 28
  • (C−R)=25−2=23(C - R) = 25 - 2 = 23
  • Denominator: (R+1)2(R+2)=32⋅4=9⋅4=36(R + 1)^2 (R + 2) = 3^2 \cdot 4 = 9 \cdot 4 = 36

Multiply the numerator terms: 806⋅28⋅23=519,064806 \cdot 28 \cdot 23 = 519{,}064

Compute the variance: Var⁡(N^C)=519,06436≈14,418.44\operatorname{Var}(\hat{N}_C) = \frac{519{,}064}{36} \approx 14{,}418.44

The standard error is the square root of the estimated variance: SE⁡(N^C)=Var⁡(N^C)=519,06436≈119–120 trout\operatorname{SE}(\hat{N}_C) = \sqrt{\operatorname{Var}(\hat{N}_C)} = \sqrt{\frac{519{,}064}{36}} \approx 119\text{--}120\text{ trout}

Rounding according to standard survey reporting yields: SE⁡(N^C)=119 trout\operatorname{SE}(\hat{N}_C) = 119\text{ trout}

Typical errors

OverviewACEpath Teacher's Diagnosis

01 The student forgets to subtract 1 at the end of Chapman's formula, overestimating N̂_C by 1.

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    N^=30⋅252  ⟹  N^=7502=375\hat{N} = \frac{30 \cdot 25}{2} \implies \hat{N} = \frac{750}{2} = 375

  2. Step 2

    N^C=(30+1)(25+1)2+1  ⟹  N^C=31⋅263  ⟹  N^C=8063≈268.67\hat{N}_C = \frac{(30 + 1)(25 + 1)}{2 + 1} \implies \hat{N}_C = \frac{31 \cdot 26}{3} \implies \hat{N}_C = \frac{806}{3} \approx 268.67

    Correct

    N^C=(M+1)(C+1)R+1−1=31⋅263−1=8063−1≈267.67≈268\hat{N}_C = \frac{(M + 1)(C + 1)}{R + 1} - 1 = \frac{31 \cdot 26}{3} - 1 = \frac{806}{3} - 1 \approx 267.67 \approx 268

    You omitted the −1- 1 term at the end of Chapman's modified estimator formula, leading to an estimate that is 1 unit too high.

  3. Step 3

    Var⁡(N^C)=31⋅26⋅(30−2)(25−2)(2+1)2(2+2)=806⋅28⋅239⋅4=51867236≈14407.56\operatorname{Var}(\hat{N}_C) = \frac{31 \cdot 26 \cdot (30 - 2)(25 - 2)}{(2 + 1)^2 (2 + 2)} = \frac{806 \cdot 28 \cdot 23}{9 \cdot 4} = \frac{518672}{36} \approx 14407.56

    Correct

    Var⁡(N^C)=(31)(26)(28)(23)36=51906436≈14418.44\operatorname{Var}(\hat{N}_C) = \frac{(31)(26)(28)(23)}{36} = \frac{519064}{36} \approx 14418.44

    You set up the variance formula correctly, but made an arithmetic multiplication error when evaluating 806⋅28⋅23806 \cdot 28 \cdot 23, getting 518672518672 instead of 519064519064.

  4. Step 4

    SE⁡(N^C)=14407.56≈120.03\operatorname{SE}(\hat{N}_C) = \sqrt{14407.56} \approx 120.03

    Correct

    SE⁡(N^C)=14418.44≈119–120\operatorname{SE}(\hat{N}_C) = \sqrt{14418.44} \approx 119\text{--}120

    You correctly took the square root of your calculated variance, but the numerical value carries forward the multiplication error from the previous step.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly identified the parameters M=30M = 30, C=25C = 25, and R=2R = 2 from the problem statement.

You accurately calculated the standard Lincoln-Petersen estimate in part (a).

You chose and correctly substituted the survey parameters into the Chapman variance formula.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Formula Omission and Arithmetic Error

You dropped the trailing −1- 1 in Chapman's formula and miscalculated the product 806⋅28⋅23806 \cdot 28 \cdot 23.

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

Chapman's estimator adds 11 to each count in the ratio to reduce positive bias when recaptures are small, but it requires subtracting 11 at the end to make the expectation unbiased (E[N^C]=NE[\hat{N}_C] = N). Omitting this final term is a common formula recall slip.

REASONING PATH
Your ACEpath teacher’s suggested approach
Check the Chapman Correction Term
  • Chapman's estimator is defined as N^C=(M+1)(C+1)R+1−1\hat{N}_C = \frac{(M + 1)(C + 1)}{R + 1} - 1.

  • Always verify that you subtract 11 after computing the quotient (M+1)(C+1)R+1\frac{(M+1)(C+1)}{R+1}.

Verify Multi-Digit Products
  • 806⋅644=519,064806 \cdot 644 = 519{,}064.

  • Break down large products into smaller components or double-check intermediate arithmetic before dividing.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Apply Chapman's complete formula: N^C=(30+1)(25+1)2+1−1=8063−1≈267.67≈268 trout\hat{N}_C = \frac{(30 + 1)(25 + 1)}{2 + 1} - 1 = \frac{806}{3} - 1 \approx 267.67 \approx 268\text{ trout}.

  2. Recalculate the numerator for variance: 806⋅28⋅23=806⋅644=519,064806 \cdot 28 \cdot 23 = 806 \cdot 644 = 519{,}064.

  3. Divide by the denominator: Var⁡(N^C)=519,06436≈14,418.44\operatorname{Var}(\hat{N}_C) = \frac{519{,}064}{36} \approx 14{,}418.44.

  4. Take the square root for the standard error: SE⁡(N^C)=14,418.44≈120.08≈119–120 trout\operatorname{SE}(\hat{N}_C) = \sqrt{14{,}418.44} \approx 120.08 \approx 119\text{--}120\text{ trout}.

Final answer

(a) N^=375 trout\hat{N} = 375\text{ trout}; (b) N^C=268 trout\hat{N}_C = 268\text{ trout}; (c) SE⁡(N^C)=119–120 trout\operatorname{SE}(\hat{N}_C) = 119\text{--}120\text{ trout}

NEXT STEPS
Final Takeaways from Your Teacher
  • Memorize the structure of Chapman's formula by pairing '+1 to each term inside' with 'subtract 1 at the end'.

  • Practice using Chapman's estimator and its variance on small-sample mark-recapture datasets to strengthen arithmetic accuracy.

OverviewACEpath Teacher's Diagnosis

02 The student confuses the standard error with the variance and omits taking the square root in part (c).

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    N^=(30⋅25)2=375\hat{N} = \frac{(30 \cdot 25)}{2} = 375

  2. Step 2

    N^C=(31⋅26)3−1  ⟹  N^C=8063−1=8033≈267.67\hat{N}_C = \frac{(31 \cdot 26)}{3} - 1 \implies \hat{N}_C = \frac{806}{3} - 1 = \frac{803}{3} \approx 267.67

  3. Step 3

    Var⁡(N^C)=31⋅26⋅28⋅2332⋅4  ⟹  Var⁡(N^C)=51867236≈14407.56\operatorname{Var}(\hat{N}_C) = \frac{31 \cdot 26 \cdot 28 \cdot 23}{3^2 \cdot 4} \implies \operatorname{Var}(\hat{N}_C) = \frac{518672}{36} \approx 14407.56

    Correct

    Var⁡(N^C)=(M+1)(C+1)(M−R)(C−R)(R+1)2(R+2)=31⋅26⋅28⋅239⋅4=519,06436≈14,418.44\operatorname{Var}(\hat{N}_C) = \frac{(M + 1)(C + 1)(M - R)(C - R)}{(R + 1)^2 (R + 2)} = \frac{31 \cdot 26 \cdot 28 \cdot 23}{9 \cdot 4} = \frac{519{,}064}{36} \approx 14{,}418.44

    You substituted the correct terms into Chapman's variance formula, but made a slight arithmetic error when multiplying the numerator terms (31⋅26⋅28⋅23=519,06431 \cdot 26 \cdot 28 \cdot 23 = 519{,}064, not 518,672518{,}672).

  4. Step 4

    SE⁡(N^C)=14407.56\operatorname{SE}(\hat{N}_C) = 14407.56

    Correct

    SE⁡(N^C)=Var⁡(N^C)=519,06436≈119 trout\operatorname{SE}(\hat{N}_C) = \sqrt{\operatorname{Var}(\hat{N}_C)} = \sqrt{\frac{519{,}064}{36}} \approx 119\text{ trout}

    You set the standard error equal to the variance rather than taking its square root.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly calculated the Lincoln-Petersen estimate for part (a).

You accurately evaluated Chapman's bias-corrected estimator for part (b).

You selected and correctly substituted into the formula for Chapman's variance in part (c).

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Omitted square root for standard error

You set SE⁡(N^C)\operatorname{SE}(\hat{N}_C) directly equal to Var⁡(N^C)\operatorname{Var}(\hat{N}_C), forgetting that standard error is the square root of the variance.

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You conflated variance with standard error/standard deviation, overlooking the dimensional discrepancy where variance is expressed in squared units (exttrout2 ext{trout}^2) while standard error must be in linear units (exttrout ext{trout}).

REASONING PATH
Your ACEpath teacher’s suggested approach
Check Units and Dimensions
  • Variance has squared units (extindividuals2 ext{individuals}^2), whereas standard error represents the dispersion on the original scale (extindividuals ext{individuals}).

  • Always convert variance to standard error by taking the principal square root: SE⁡=Var⁡\operatorname{SE} = \sqrt{\operatorname{Var}}.

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Recalculate the numerator of the variance: 31⋅26⋅28⋅23=806⋅644=519,06431 \cdot 26 \cdot 28 \cdot 23 = 806 \cdot 644 = 519{,}064.

  2. Divide by the denominator: Var⁡(N^C)=519,06436≈14,418.44\operatorname{Var}(\hat{N}_C) = \frac{519{,}064}{36} \approx 14{,}418.44.

  3. Take the square root to determine the standard error: SE⁡(N^C)=14,418.44≈120.08 trout\operatorname{SE}(\hat{N}_C) = \sqrt{14{,}418.44} \approx 120.08\text{ trout} (or ≈119 trout\approx 119\text{ trout} using exact rational rounding).

Final answer

SE⁡(N^C)≈119 trout\operatorname{SE}(\hat{N}_C) \approx 119\text{ trout} (or 120 trout120\text{ trout})

NEXT STEPS
Final Takeaways from Your Teacher
  • Double-check long multiplication steps or use factor cancellation before multiplying large numbers.

  • Label statistical units explicitly at each stage to remind yourself when a square root is required.

OverviewACEpath Teacher's Diagnosis

03 The student neglects the square on (R + 1) in the denominator of the variance formula, dividing by (R + 1)(R + 2) instead of (R + 1)²(R + 2).

Sample Student Work

Reviewed by ACEpath Teacher
Step-by-Step Review
Walk Through It with Your Teacher
Partially correct
  1. Step 1

    N^=(30⋅25)2=375\hat{N} = \frac{(30 \cdot 25)}{2} = 375

  2. Step 2

    N^C=(31⋅26)3−1=8033≈267.67\hat{N}_C = \frac{(31 \cdot 26)}{3} - 1 = \frac{803}{3} \approx 267.67

  3. Step 3

    Var⁡(N^C)=31⋅26⋅28⋅23(2+1)(2+2)  ⟹  Var⁡(N^C)=51867212≈43222.67\operatorname{Var}(\hat{N}_C) = \frac{31 \cdot 26 \cdot 28 \cdot 23}{(2+1)(2+2)} \implies \operatorname{Var}(\hat{N}_C) = \frac{518672}{12} \approx 43222.67

    Correct

    Var⁡(N^C)=(M+1)(C+1)(M−R)(C−R)(R+1)2(R+2)=31⋅26⋅28⋅2332⋅4=51906436≈14418.44\operatorname{Var}(\hat{N}_C) = \frac{(M + 1)(C + 1)(M - R)(C - R)}{(R + 1)^2(R + 2)} = \frac{31 \cdot 26 \cdot 28 \cdot 23}{3^2 \cdot 4} = \frac{519064}{36} \approx 14418.44

    You omitted the square on (R+1)(R + 1) in the denominator, evaluating (2+1)(2+2)=12(2 + 1)(2 + 2) = 12 instead of (2+1)2(2+2)=36(2 + 1)^2(2 + 2) = 36.

  4. Step 4

    SE⁡(N^C)=43222.67≈207.90\operatorname{SE}(\hat{N}_C) = \sqrt{43222.67} \approx 207.90

    Correct

    SE⁡(N^C)=Var⁡(N^C)=14418.44≈119–120 trout\operatorname{SE}(\hat{N}_C) = \sqrt{\operatorname{Var}(\hat{N}_C)} = \sqrt{14418.44} \approx 119\text{--}120\text{ trout}

    You took the square root correctly, but the resulting standard error was inflated due to the denominator error in the variance step.

WHAT YOU DID WELL
Compliments from Your Teacher

You correctly computed the standard Lincoln-Petersen estimate.

You accurately applied Chapman's estimator and handled the subtraction of 1.

THE ERROR
Where Your ACEpath Teacher Spotted the Mistake
Omitted Exponent in Variance Denominator

You wrote (2+1)(2+2)(2+1)(2+2) in the denominator instead of (2+1)2(2+2)(2+1)^2(2+2).

WHY IT HAPPENED
ACEpath Teacher's Diagnosis

You overlooked the power of 2 on the (R+1)(R + 1) term in Chapman's variance formula, dividing by (R+1)(R+2)(R + 1)(R + 2) rather than (R+1)2(R+2)(R + 1)^2(R + 2).

REASONING PATH
Your ACEpath teacher’s suggested approach
Chapman Variance Formula Structure
  • Var⁡(N^C)=(M+1)(C+1)(M−R)(C−R)(R+1)2(R+2)\operatorname{Var}(\hat{N}_C) = \frac{(M + 1)(C + 1)(M - R)(C - R)}{(R + 1)^2(R + 2)}

  • Double-check that the first factor in the denominator is squared: (R+1)2(R + 1)^2, not (R+1)(R + 1).

CORRECTED PATH
ACEpath Teacher's Demonstration
  1. Compute the denominator: (R+1)2(R+2)=32⋅4=9⋅4=36(R + 1)^2 (R + 2) = 3^2 \cdot 4 = 9 \cdot 4 = 36.

  2. Compute the numerator: 31⋅26⋅28⋅23=806⋅644=51906431 \cdot 26 \cdot 28 \cdot 23 = 806 \cdot 644 = 519064.

  3. Calculate the variance: Var⁡(N^C)=51906436≈14418.44\operatorname{Var}(\hat{N}_C) = \frac{519064}{36} \approx 14418.44.

  4. Take the square root to find the standard error: SE⁡(N^C)=14418.44≈120 trout\operatorname{SE}(\hat{N}_C) = \sqrt{14418.44} \approx 120\text{ trout} (or 119119 depending on rounding).

Final answer

(a) N^=375 trout\hat{N} = 375\text{ trout}; (b) N^C=268 trout\hat{N}_C = 268\text{ trout}; (c) SE⁡(N^C)=119–120 trout\operatorname{SE}(\hat{N}_C) = 119\text{--}120\text{ trout}

NEXT STEPS
Final Takeaways from Your Teacher
  • Write down (R+1)2(R + 1)^2 explicitly as (R+1)(R+1)(R + 1)(R + 1) during scratch work to avoid dropping the square.

  • Check whether your variance denominator has three total linear factors of RR terms: (R+1)(R+1), (R+1)(R+1), and (R+2)(R+2).

Lock it in

Similar question

01

A wildlife biologist marks M=19M = 19 salamanders in a wetland. In a second survey, C=11C = 11 salamanders are caught, of which R=3R = 3 are marked. Calculate Chapman's estimate N^C\hat{N}_C.

Answer

N^C=(19+1)(11+1)3+1−1=20⋅124−1=60−1=59 salamanders\hat{N}_C = \frac{(19 + 1)(11 + 1)}{3 + 1} - 1 = \frac{20 \cdot 12}{4} - 1 = 60 - 1 = 59\text{ salamanders}.

02

In a study of an endangered lizard, M=15M = 15 individuals are marked. Later, C=17C = 17 lizards are captured, with R=2R = 2 marked recaptures. Calculate (a) the Lincoln-Petersen estimate N^\hat{N} and (b) Chapman's estimate N^C\hat{N}_C, then compare the difference.

Answer

(a) N^=15⋅172=127.5≈128\hat{N} = \frac{15 \cdot 17}{2} = 127.5 \approx 128; (b) N^C=(16)(18)3−1=96−1=95\hat{N}_C = \frac{(16)(18)}{3} - 1 = 96 - 1 = 95. The Lincoln-Petersen estimator overestimates the population by about 3333 lizards due to small-sample bias.

03

An ecologist tags M=40M = 40 songbirds. A secondary netting yields C=30C = 30 birds with R=4R = 4 recaptures. Calculate Chapman's estimate N^C\hat{N}_C and its estimated variance Var⁡(N^C)=(M+1)(C+1)(M−R)(C−R)(R+1)2(R+2)\operatorname{Var}(\hat{N}_C) = \frac{(M + 1)(C + 1)(M - R)(C - R)}{(R + 1)^2 (R + 2)}.

Answer

N^C=(41)(31)5−1=12715−1=254.2−1≈253 birds\hat{N}_C = \frac{(41)(31)}{5} - 1 = \frac{1271}{5} - 1 = 254.2 - 1 \approx 253\text{ birds}. Var⁡(N^C)=(41)(31)(36)(26)52⋅6=1271⋅93625⋅6=1,189,656150≈7931.04\operatorname{Var}(\hat{N}_C) = \frac{(41)(31)(36)(26)}{5^2 \cdot 6} = \frac{1271 \cdot 936}{25 \cdot 6} = \frac{1{,}189{,}656}{150} \approx 7931.04.

Loading workspace…