Resources
A comparative morphologist analyzes five vertebrate taxa (, , , , and ) scored for five binary synapomorphies (coded as = ancestral, = derived): dorsal nerve cord, paired appendages, lungs, keratinized scales, and endothermy. The character state distributions are: = ; = ; = ; = ; and = . Rather than determining sister taxa, calculate the Jaccard similarity coefficient () between taxon and taxon , defined as , where is the number of shared derived states (), is the number of characters where only has the derived state, and is the number of characters where only has the derived state.
Answer and steps
Final answer
(or ).
Steps
Step 1: Identify the Character States for Taxon and Taxon
From the data provided, compare the character states position by position for taxon and taxon :
- Taxon :
- Taxon :
Listing the pairs of states for each of the five characters:
- Dorsal nerve cord:
- Paired appendages:
- Lungs:
- Keratinized scales:
- Endothermy:
Step 2: Count the Components of the Jaccard Formula
Using the definitions provided in the question:
- is the number of shared derived characters where both taxa have state :
- is the number of characters where taxon has and taxon has :
- is the number of characters where taxon has and taxon has :
(Note: Shared ancestral absences, denoted as , are intentionally excluded from the Jaccard similarity coefficient to avoid grouping taxa solely by the lack of derived features.)
Step 3: Calculate the Jaccard Similarity Coefficient ()
Substitute the counts into the formula:
Typical errors
01 The student confuses the Jaccard similarity coefficient with the simple matching coefficient by including shared ancestral states (M00) in the denominator.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
X = (1, 1, 1, 0, 0) ⟹ Z = (1, 1, 1, 1, 1)
- Step 2
M_{11} = 3, M_{01} = 2, M_{10} = 0, M_{00} = 0
- Step 3
= 2 + 0 + 3 + 2 = 7
Correct= M_{01} + M_{10} + M_{11} = 2 + 0 + 3 = 5_
You added a fourth term of to the denominator. The Jaccard coefficient formula strictly excludes shared ancestral absences (), defining the denominator as . Furthermore, there are zero characters where both taxa share state (), so substituting counted the ancestral states of taxon rather than shared absences.
- Step 4
J =
CorrectJ = = = 0.600
Because your denominator was incorrectly calculated as , your final similarity coefficient propagated this error.
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly transcribed the character state vectors for both taxa.
You accurately identified that taxon and taxon share three derived states ().
You correctly identified the two characters where taxon has the derived state and taxon has the ancestral state ().
THE ERROR
Where Your ACEpath Teacher Spotted the MistakeYou substituted an extra term of into the denominator sum, confusing the Jaccard denominator with the Simple Matching Coefficient denominator and miscounting the ancestral characters.
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou included shared ancestral absence () in the denominator of the Jaccard index—which only considers characters present in at least one taxon ()—and you mistook the two ancestral zeros in taxon for mutual shared absences (), even though taxon possesses derived states () at those positions.
REASONING PATH
Your ACEpath teacher’s suggested approach- Jaccard vs. Simple Matching Coefficient
The Jaccard index ignores mutual absences: , whereas the Simple Matching Coefficient includes them: .
Always verify whether the requested similarity metric excludes pairings before setting up your denominator.
- Pairwise Matching of Shared Absences
counts only positions where both taxa display simultaneously: .
Do not count zeros from a single taxon; compare both taxa position by position to ensure both are .
CORRECTED PATH
ACEpath Teacher's DemonstrationWrite the Jaccard denominator containing only characters present in at least one taxon: .
Substitute the counts obtained from Step 2: .
Compute the ratio: .
J = = 0.600 60.0%
02 The student counts the total number of characters across both taxa or adds the dimensions incorrectly in the denominator, taking total 1s as 3 + 5 = 8.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
,
- Step 2Correct
,
You added the total count of ones in taxon () and taxon (). In the Jaccard denominator, each character position is counted at most once if either taxon has a derived state, giving , , and for a sum of .
- Step 3Correct
Because of the miscalculated denominator, your final Jaccard similarity coefficient evaluated to instead of .
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly recorded the character state vectors for both taxa.
You accurately identified the number of shared derived character states as .
THE ERROR
Where Your ACEpath Teacher Spotted the Mistakeno
WHY IT HAPPENED
ACEpath Teacher's DiagnosisYou treated the Jaccard denominator as the arithmetic sum of derived states in both taxa () instead of the union of derived states (), which double-counts the shared characters ().
REASONING PATH
Your ACEpath teacher’s suggested approach- Union vs. Sum of Sizes
The Jaccard denominator represents the total number of characters where at least one taxon displays state : .
Categorize each character position individually into , , , or before summing the categories.
CORRECTED PATH
ACEpath Teacher's DemonstrationCount the remaining non-zero character pairings: (characters 4 and 5) and .
Calculate the denominator: .
Compute the ratio: .
03 The student swaps taxa X and Y, using Y = (1, 1, 1, 1, 0) instead of X when comparing against Z.
Sample Student Work
Reviewed by ACEpath TeacherStep-by-Step Review
Walk Through It with Your Teacher- Step 1
Comparing and
CorrectTaxon and Taxon
You extracted the vector for taxon , , instead of the vector given for taxon , .
- Step 2Correct
Your tallying logic is completely accurate for the vectors you wrote, but the counts were shifted because of the swapped taxon vector in Step 1.
- Step 3Correct
You correctly substituted your counts into the Jaccard index formula, but the final numerical value is affected by the initial taxon error.
WHAT YOU DID WELL
Compliments from Your TeacherYou correctly identified and applied the formula for the Jaccard similarity coefficient.
You accurately tallied the paired binary character states for the two vectors you wrote down.
THE ERROR
Where Your ACEpath Teacher Spotted the MistakeYou accidentally pulled the vector for taxon instead of taxon .
WHY IT HAPPENED
ACEpath Teacher's DiagnosisA transcription and label-tracking slip occurred when scanning the nested list of consecutive vertebrate taxa, grabbing row instead of row .
REASONING PATH
Your ACEpath teacher’s suggested approach- Data Extraction Check
Always write the taxon label directly next to its extracted vector before computing pairwise comparisons.
Explicitly label and on your scratch paper to verify alignment against the problem statement.
CORRECTED PATH
ACEpath Teacher's DemonstrationIdentify the character states for taxon and taxon .
Count the pairwise character states: characters 1, 2, and 3 are so ; characters 4 and 5 are so ; no characters are so .
Substitute these counts into the Jaccard index formula: .
Lock it in
What is the formula for the Jaccard similarity coefficient () between two taxa with binary character states?
. Mutual absences () are completely excluded from both the numerator and the denominator.
Similar question
Two taxa, and , have binary character vectors and . Calculate the Jaccard similarity coefficient () between and .
Answer
. Matching pairs position-by-position: position 1 is ; position 2 is ; position 3 is ; position 4 is . The Jaccard coefficient excludes : .
Consider taxa and . Compute both the Jaccard similarity coefficient () and the Simple Matching Coefficient ().
Answer
and . Pairwise comparisons: at positions 1 and 4 (); at position 2 (); at position 5 (); at positions 3 and 6 (). Then , while .
A phylogenetic dataset assesses six binary characters across three taxa: , , and . Identify which pair of taxa (-, -, or -) shares the highest Jaccard similarity coefficient, and calculate that maximum value.
Answer
Pair - has the highest coefficient with . For and : (chars 3, 5), (chars 1, 6), . For and : (chars 1, 3, 6), (char 5), (char 2) J(T_1, T_3) = = = 0.600T_2T_3M_{11} = 1M_{10} = 1M_{01} = 3._