Data Reference

Complete reference for all question groups, score types, weights, and preprocessing rules used by the eTeamUps V1 matching algorithm.

Question Groups

The algorithm uses 19 question groups. Of these, 16 are scored and contribute to the overall match. Two are excluded from the core set entirely (QG6, QG19), and one is in the core set but skipped because its score type is unimplemented (QG3).

ID Name Score Type Questions Weight Status
1 values TOP_TEN 21 100 Scored
2 teamwork MAX_MEAN 10 100 Scored
3 pacing-v1 INDIVIDUAL_SUPPLEMENTARY 4 100 Skipped (unimplemented)
4 workexpectation SUPPLEMENTARY 4 100 Scored
5 reliance MIN_MEAN 7 100 Scored
6 conflict-v1 CONFLICT 4 Not in core set
7 goalpriority MAX_MEAN 6 100 Scored
8 taskrelation SUPPLEMENTARY 11 100 Scored
9 timeurgency SUPPLEMENTARY 7 100 Scored
10 workethic SUPPLEMENTARY 7 100 Scored
11 conscientiousness MAX_MEAN 11 100 Scored
12 learnorientation MAX_MEAN 7 50 Scored
13 assertiveness COMPLEMENTARY 11 100 Scored
14 agreeableness MAX_MEAN 11 100 Scored
15 performanceprove MIN_MEAN 6 50 Scored
16 performanceavoid MIN_MEAN 6 50 Scored
17 intellect MAX_MEAN 8 100 Scored
18 conflict-v2 GROUP_AVERAGE 16 50 Scored
19 pacing-v2 SUPPLEMENTARY 1 Not in core set

Total: 158 questions across 19 groups. 16 groups scored with a combined weight pool of 1,350.


Score Types

Score Type Calculator What it Measures Higher Score Means
SUPPLEMENTARY createSupplementaryCalculatorV1 Team homogeneity (low stddev) Candidate makes the team more similar
COMPLEMENTARY createComplementaryCalculatorV1 Team diversity (high stddev) Candidate increases team diversity
MAX_MEAN createMeanMaxCalculatorV1 Trait strength (high mean) Adding the candidate raises the team average
MIN_MEAN createMeanMinCalculatorV1 Trait absence (low mean) Adding the candidate lowers the team average
TOP_TEN createValuesCalculatorV1 Value overlap (shared top answers) Candidate shares more top values with the team
GROUP_AVERAGE createConflictCalculatorV1 Conflict style (sub-group scoring) Team + candidate have better conflict styles

Score Type Formulas Summary

Type Candidate Score Formula
SUPPLEMENTARY 100 - (sampleStdDev(teamMeans + candidateMean) / theoreticalMaxStdDev * 100)
COMPLEMENTARY sampleStdDev(teamMeans + candidateMean) / theoreticalMaxStdDev * 100
MAX_MEAN (mean(teamMeans + candidateMean) - 1) / 4 * 100
MIN_MEAN 100 - ((mean(teamMeans + candidateMean) - 1) / 4 * 100)
TOP_TEN (rawOverlapScore - minPossible) / (maxPossible - minPossible) * 100
GROUP_AVERAGE sum(teamConflictScores + candidateScore) / ((teamSize + 1) * 13) * 100

Score Categories

Score (rounded) Category
0 – 32 LOW
33 – 65 MEDIUM
66 – 100 HIGH

Dimension Details

Values (QG1) – TOP_TEN

21 questions using the “important” Likert scale (1=Not Important to 5=Very Important).

Measures alignment of personal values. Each person’s top 5 values (answers > 3, sorted by value then question ID) are compared. More overlap with the team’s top values = higher score.

Questions (Q1-Q21):

Q# Value
Q1 Reliability (others can rely on me)
Q2 Honesty (direct communication)
Q3 Harmony (working without conflict)
Q4 Fame (name recognition)
Q5 Environment (natural environment concern)
Q6 Ethics (doing things ethically)
Q7 Autonomy (doing what I want)
Q8 Fun (having fun, feeling good)
Q9 Creativity (opportunity to be creative)
Q10 Money (making money on a project)
Q11 Innovation (creating something new)
Q12 Acknowledgment (recognized for contributions)
Q13 Influence (influencing others)
Q14 Process (taking pleasure in the process)
Q15 Results (getting results)
Q16 Collaboration (working collaboratively)
Q17 Ahead of schedule
Q18 Deadlines (meeting deadlines)
Q19 Long-term success
Q20 Short-term success
Q21 Hard work (working hard)

Teamwork (QG2) – MAX_MEAN

10 questions, “agree” Likert scale (1=Strongly Disagree to 5=Strongly Agree).

Higher mean = more team-oriented.

  • Faking: Q26 (“Working in teams is always the most effective way to work”)
  • Reverse scored: Q25, Q27, Q31 (autonomy-preference questions)

Work Expectation (QG4) – SUPPLEMENTARY

4 questions, “agree” Likert scale. Measures expectations around work hours and availability.

  • Reverse scored: Q38 (“I expect team members to be available 24/7”)

Similar expectations within a team = higher score.


Reliance (QG5) – MIN_MEAN

7 questions, “agree” Likert scale. Measures dependence on others.

  • Faking: Q41 (“I always trust my team members to make good contributions”)
  • Reverse scored: Q44, Q45, Q46

Lower mean = more self-sufficient = higher score.


Goal Priority (QG7) – MAX_MEAN

6 questions, “agree” Likert scale. Measures team-goal orientation vs. self-interest.

  • Faking: Q54 (“Nothing should come before the goals of the team”)
  • Reverse scored: Q55, Q56

Higher mean = puts team goals first.


Task Relation (QG8) – SUPPLEMENTARY

11 questions, “agree” Likert scale. Measures task-focus vs. relationship orientation.

  • Faking: Q61 (“I always remember people’s names”)
  • Reverse scored: Q63, Q64, Q65

Similar orientation within team = higher score.


Time Urgency (QG9) – SUPPLEMENTARY

7 questions, “agree” Likert scale. Measures pace preference and time pressure.

  • Reverse scored: Q69 (“I typically work carefully and deliberately”)

Similar pace within team = higher score.


Work Ethic (QG10) – SUPPLEMENTARY

7 questions, “agree” Likert scale. Measures drive and effort.

No faking or reverse-scored questions.

Similar work ethic within team = higher score.


Conscientiousness (QG11) – MAX_MEAN

11 questions, “agree” Likert scale. Measures attention to detail and follow-through.

  • Faking: Q87 (“I’m never late to appointments”)

Higher mean = more conscientious = better.


Learn Orientation (QG12) – MAX_MEAN

7 questions, “agree” Likert scale. Measures desire to learn and grow.

  • Faking: Q99 (“I believe learning through work assignments is the only way to learn”)

Higher mean = stronger learning drive.

Weight: 50 (half weight).


Assertiveness (QG13) – COMPLEMENTARY

11 questions, “agree” Likert scale. Measures dominance vs. deference.

No faking or reverse-scored questions.

Q100-Q105: high-assertiveness items. Q106-Q110: low-assertiveness items.

Higher diversity in assertiveness = better (COMPLEMENTARY scoring). Teams benefit from a mix of assertive and deferential members.


Agreeableness (QG14) – MAX_MEAN

11 questions, “agree” Likert scale. Measures empathy, respect, and cooperation.

  • Faking: Q113 (“I never get in arguments”)

Higher mean = more agreeable = better.


Performance Prove (QG15) – MIN_MEAN

6 questions, “agree” Likert scale. Measures need to prove competence to others.

Lower mean = less need to show off = better.

Weight: 50 (half weight).


Performance Avoid (QG16) – MIN_MEAN

6 questions, “agree” Likert scale. Measures tendency to avoid challenging tasks.

  • Faking: Q131 (“I always take on assignments even if they are too complex for me”)

Lower mean = less avoidance = better.

Weight: 50 (half weight).


Intellect (QG17) – MAX_MEAN

8 questions, “agree” Likert scale. Measures intellectual curiosity and capability.

Higher mean = higher intellect = better.


Conflict v2 (QG18) – GROUP_AVERAGE

16 questions, “usually” Likert scale (1=Never to 5=Always). Questions are grouped by conflict sub-type via the metadata field.

Sub-group Questions Conflict Score
ConflictYield Q142, Q143, Q144, Q145 0
ConflictForce Q146, Q147, Q148, Q149 0
ConflictComp Q150, Q151 7
ConflictPS Q152, Q153, Q154 13
ConflictAvoid Q155, Q156, Q157 0

Each person’s dominant style is the sub-group with the highest sum (all sub-group sums are compared using the same denominator – the total QG18 question count – so it is effectively a raw-sum comparison). Only Problem-Solving (13) and Compromise (7) contribute positive scores.

Weight: 50 (half weight).


Faking Questions

These 8 questions are excluded from all scoring. They use extreme wording that no honest respondent would strongly agree with.

Q# Group Text
Q26 QG2 (Teamwork) Working in teams is always the most effective way to work
Q41 QG5 (Reliance) I always trust my team members to make good contributions
Q54 QG7 (Goal Priority) Nothing should come before the goals of the team
Q61 QG8 (Task Relation) I always remember people’s names
Q87 QG11 (Conscientiousness) I’m never late to appointments
Q99 QG12 (Learn Orientation) I believe learning through work assignments is the only way to learn
Q113 QG14 (Agreeableness) I never get in arguments
Q131 QG16 (Perf. Avoid) I always take on assignments even if they are too complex for me

:::tip
Faking questions typically contain absolute words like “always”, “never”, or “only”. They are included in the survey but silently dropped during scoring.
:::


Reverse-Scored Questions

These questions are phrased opposite to the dimension’s direction. Their raw values are flipped before scoring using the reverse map:

Raw Effective
1 5
2 4
3 3
4 2
5 1
Q# Group Text
Q25 QG2 (Teamwork) I like a good deal of autonomy when I work
Q27 QG2 (Teamwork) I’m most productive when people let me do my own thing
Q31 QG2 (Teamwork) I like to do things my own way
Q38 QG4 (Work Expectation) I expect team members to be available 24/7
Q44 QG5 (Reliance) I try to minimize the extent that I have to rely on my team members
Q45 QG5 (Reliance) I like to check over teammates work to make sure they have done a good job
Q46 QG5 (Reliance) A team member must earn their keep before I can rely on them
Q55 QG7 (Goal Priority) My personal goals should be given priority over team goals
Q56 QG7 (Goal Priority) I don’t want to be part of a team if it doesn’t help me achieve my personal goals
Q63 QG8 (Task Relation) I prefer to stay on task during team meetings
Q64 QG8 (Task Relation) Sticking to an agenda during a team meeting is important to me
Q65 QG8 (Task Relation) I get frustrated when people bring up their personal issues during team meetings
Q69 QG9 (Time Urgency) I typically work carefully and deliberately

Production Weights

Weights come from the QuestionGroupSetGroup table and determine each dimension’s contribution to the overall score.

Weight Dimensions
100 Values, Teamwork, Work Expectation, Reliance, Goal Priority, Task Relation, Time Urgency, Work Ethic, Conscientiousness, Assertiveness, Agreeableness, Intellect
50 Learn Orientation, Performance Prove, Performance Avoid, Conflict v2

Combined weight pool: (12 * 100) + (4 * 50) = 1,400

But since Pacing v1 (QG3, weight 100) is skipped, the effective pool is:
(11 * 100) + (4 * 50) = 1,300

Wait – let me recalculate. The 16 scored dimensions:

  • Weight 100 (12 groups): QG1, QG2, QG4, QG5, QG7, QG8, QG9, QG10, QG11, QG13, QG14, QG17
  • Weight 50 (4 groups): QG12, QG15, QG16, QG18

Effective weight pool: (12 * 100) + (4 * 50) = 1,400


Likert Answer Sets

Questions use one of three Likert scales:

Set Scale Used By
important 1=Not Important, 2=Slightly Important, 3=Moderately Important, 4=Important, 5=Very Important QG1 (Values)
agree 1=Strongly Disagree, 2=Disagree, 3=Neutral, 4=Agree, 5=Strongly Agree QG2, QG4-QG17
usually 1=Never, 2=Rarely, 3=Sometimes, 4=Usually, 5=Always QG18 (Conflict v2)
(none) Context-dependent QG3, QG6, QG19 (not scored)

Skipped Score Types

Two score types are declared in the enum but not implemented in V1:

Score Type Group Reason
INDIVIDUAL_SUPPLEMENTARY QG3 (Pacing v1) The Java implementation throws UnsupportedOperationException. Skipped in scoring.
CONFLICT QG6 (Conflict v1) Superseded by QG18 (Conflict v2, GROUP_AVERAGE). Not in the core question group set.