How to Create Random Pairs From Two Lists
To create random pairs from two lists, verify the roles and counts, shuffle one list, and match it position by position to the other. With 5 mentors and 5 learners, each named mentor has probability 1/5 of being matched to any particular learner if the learner list is shuffled uniformly.
A worked example with numbers
Five mentors M1–M5 and five learners L1–L5 need one-to-one matches. Keep mentor order fixed, shuffle learners into L3, L1, L5, L2, L4, then pair M1–L3 through M5–L4. M1 has 5 possible learners, so P(M1 with L3) = 1/5 = 20%. The output has five pairs and no mentor or learner repeats.
How to set the rule before the result
Confirm that each list contains one role and equal counts. A list shuffle can randomise one side; the one-list pair tool is not a substitute because it mixes roles into one roster.
Common mistakes that change the odds or the process
Do not concatenate two role lists and pair adjacent entries: it can produce mentor–mentor or learner–learner pairs. Do not silently discard an extra entry when counts differ. Do not randomise prohibited matches after the fact; record exclusions before using a constraint-aware process.
Where this method stops being appropriate
This simple method assumes every cross-list pairing is allowed. Capacity limits, safeguarding, language needs, conflicts, and ranked preferences are constraints that require a designed matching process rather than a blind shuffle.
How the random source fits into the rule
NIST’s hypergeometric-distribution glossary
Apply the rule to the actual input
Confirm that each list contains one role and equal counts. A list shuffle can randomise one side; the one-list pair tool is not a substitute because it mixes roles into one roster.
Audit the calculation or allocation
Five mentors M1–M5 and five learners L1–L5 need one-to-one matches. Keep mentor order fixed, shuffle learners into L3, L1, L5, L2, L4, then pair M1–L3 through M5–L4. M1 has 5 possible learners, so P(M1 with L3) = 1/5 = 20%. The output has five pairs and no mentor or learner repeats.
Do not import a different rule by accident
Do not concatenate two role lists and pair adjacent entries: it can produce mentor–mentor or learner–learner pairs. Do not silently discard an extra entry when counts differ. Do not randomise prohibited matches after the fact; record exclusions before using a constraint-aware process.
Limit of this specific method
This simple method assumes every cross-list pairing is allowed. Capacity limits, safeguarding, language needs, conflicts, and ranked preferences are constraints that require a designed matching process rather than a blind shuffle.
Source and reproducibility
NIST’s hypergeometric-distribution glossary
Validate the two-role matrix before shuffling
| Mentors | Learners | One-to-one pairing possible? |
|---|---|---|
| 5 | 5 | Yes |
| 5 | 4 | No; define one unpaired mentor |
| 5 | 6 | No; define one unpaired learner |
Equal counts are a capacity fact, not a random result. If one role has fewer entries, decide whether somebody observes, receives a second assignment, or waits for another round. A shuffle cannot create a partner that is absent.
Keep constraints in the input, not as a redraw excuse
If M2 cannot work with L4, a simple uniform permutation may yield an invalid pair. Listing prohibited edges before matching is the honest rule; a constraint-aware matcher may then sample from allowed assignments. Repeatedly shuffling until a pleasant-looking result appears can change probabilities among the allowed pairs unless the acceptance method is designed and disclosed. For consequential mentoring, human suitability should be evaluated before any tie-breaking randomisation.
Reproduce this result before relying on it
Bipartite random pairing matches each entry in one role-specific list to one entry in a different role-specific list. Five mentors M1–M5 and five learners L1–L5 need one-to-one matches. Keep mentor order fixed, shuffle learners into L3, L1, L5, L2, L4, then pair M1–L3 through M5–L4. M1 has 5 possible learners, so P(M1 with L3) = 1/5 = 20%. The output has five pairs and no mentor or learner repeats.
Choose the action that matches the stated rule
Confirm that each list contains one role and equal counts. A list shuffle can randomise one side; the one-list pair tool is not a substitute because it mixes roles into one roster. Do not concatenate two role lists and pair adjacent entries: it can produce mentor–mentor or learner–learner pairs. Do not silently discard an extra entry when counts differ. Do not randomise prohibited matches after the fact; record exclusions before using a constraint-aware process.
What the number does not decide
This simple method assumes every cross-list pairing is allowed. Capacity limits, safeguarding, language needs, conflicts, and ranked preferences are constraints that require a designed matching process rather than a blind shuffle. NIST’s hypergeometric-distribution glossary
State the complete decision model
To create random pairs from two lists, verify the roles and counts, shuffle one list, and match it position by position to the other. With 5 mentors and 5 learners, each named mentor has probability 1/5 of being matched to any particular learner if the learner list is shuffled uniformly. Bipartite random pairing matches each entry in one role-specific list to one entry in a different role-specific list.
Before publishing or using the outcome
Five mentors M1–M5 and five learners L1–L5 need one-to-one matches. Keep mentor order fixed, shuffle learners into L3, L1, L5, L2, L4, then pair M1–L3 through M5–L4. M1 has 5 possible learners, so P(M1 with L3) = 1/5 = 20%. The output has five pairs and no mentor or learner repeats. Confirm that each list contains one role and equal counts. A list shuffle can randomise one side; the one-list pair tool is not a substitute because it mixes roles into one roster.
Final rule check for this allocation
To create random pairs from two lists, verify the roles and counts, shuffle one list, and match it position by position to the other. With 5 mentors and 5 learners, each named mentor has probability 1/5 of being matched to any particular learner if the learner list is shuffled uniformly. Bipartite random pairing matches each entry in one role-specific list to one entry in a different role-specific list. This simple method assumes every cross-list pairing is allowed. Capacity limits, safeguarding, language needs, conflicts, and ranked preferences are constraints that require a designed matching process rather than a blind shuffle.
Verify one-to-one coverage after matching
Read both columns after the match: each mentor should occur once and each learner should occur once. With five entries in each role, the result must contain five rows, not four or six. This simple count catches a copied row and a missed person without judging whether the particular random pairing is liked. If a participant withdraws, apply the stated replacement policy rather than editing only the visible pair.