Measure What Matters

The extent to which a measure captures the intended construct for a specified purpose; the overall strength of relationship between observed variable and latent concept. (Measure only what changes the decision.)

started 0 · finished 0 (claimed) · not yet measured (verified) · no data (n<5)

Derived from: Measure What MattersT1

Ordered tasks (22) — this is what a project auto-creates

#1Construct Definition & Decision ClarityHighTo Do
Pin down exactly what you intend to measure and tie that definition to the decision it must inform, before you write a single item.
Assignee: Unassigned · due 0 days after project start · takes 14 days
#2Item / Question Design QualityHighTo Do
Word, structure, and format individual items so respondents interpret them as you intend and can answer accurately.
Assignee: Unassigned · due 14 days after project start · takes 14 days
#3Scale / Measurement Level ChoiceMediumTo Do
Tells you how to fix the measurement level — nominal, ordinal, interval, or ratio — for each thing you count, and which arithmetic that level actually licenses.
Assignee: Unassigned · due 28 days after project start · takes 7 days
#4Content Sampling & Indicator AdequacyMediumTo Do
Choose a set of indicators that covers the whole content domain of your construct without padding it with irrelevant or redundant items.
Assignee: Unassigned · due 35 days after project start · takes 7 days
#5Sample Adequacy & RepresentativenessMediumTo Do
How large and how representative your development sample must be for factor structures, reliability estimates, and item statistics to be stable rather than accidents of your particular respondents.
Assignee: Unassigned · due 42 days after project start · takes 7 days
#6Mode & Contextual ConditionsMediumTo Do
Address the administration conditions you inherit rather than design—mode, medium, question sensitivity, continuous-testing pressure—and how they reshape how respondents answer.
Assignee: Unassigned · due 49 days after project start · takes 7 days
#7Respondent Cognitive Response ProcessHighTo Do
Unpacks the four mental steps—comprehension, retrieval, judgment, response mapping—every respondent runs to answer a question, and where each step breaks.
Assignee: Unassigned · due 56 days after project start · takes 14 days
#8Respondent Ability, Willingness & MotivationMediumTo Do
Separates the two distinct reasons a respondent gives a poor answer—they can't (ability) or they won't (willingness)—and what each requires.
Assignee: Unassigned · due 70 days after project start · takes 7 days
#9Response Rate & Data CompletenessMediumTo Do
How much of your intended sample actually delivered usable data, and why the missing part matters more than the count that responded.
Assignee: Unassigned · due 77 days after project start · takes 7 days
#10Probability Sampling DesignLowTo Do
How the probabilistic mechanism you use to select respondents — stratifying, clustering, sampling in stages — governs the precision and bias of any population estimate your measure feeds into.
Assignee: Unassigned · due 84 days after project start · takes 3 days
#11Sampling Frame QualityLowTo Do
Audit the list you actually sample from—before it silently distorts every estimate downstream.
Assignee: Unassigned · due 87 days after project start · takes 3 days
#12Data Screening & Analytic Procedure RigorMediumTo Do
The pre-analysis and modeling decisions—cleaning, extraction, retention, rotation, estimation—that determine whether your factor structure reflects the data or your defaults.
Assignee: Unassigned · due 90 days after project start · takes 7 days
#13Random Measurement ErrorHighTo Do
The noise floor of any measurement—the inconsistent scatter that shows up when you measure the same thing twice.
Assignee: Unassigned · due 97 days after project start · takes 14 days
#14Systematic Error & BiasHighTo Do
Address the persistent offsets that make a metric consistently point at the wrong target—biases that repetition will not average away.
Assignee: Unassigned · due 111 days after project start · takes 14 days
#15ReliabilityHighTo Do
Define how consistently your instrument reproduces results and how to quantify that consistency as a ratio of true signal to observed variation.
Assignee: Unassigned · due 125 days after project start · takes 14 days
#16Validity & Total Measurement QualityHighTo Do
Whether your measure actually captures the concept you care about for the decision at hand—the overall verdict on measurement quality.
Assignee: Unassigned · due 139 days after project start · takes 14 days
#17Dimensionality & Factor StructureMediumTo Do
Check whether the items you are combining actually reflect one underlying thing—or several—before you sum them into a score.
Assignee: Unassigned · due 153 days after project start · takes 7 days
#18Comparability, Equating & Test SecurityMediumTo Do
Address whether scores from different forms, times, languages, or channels mean the same thing—and how to protect the instrument so they keep meaning the same.
Assignee: Unassigned · due 160 days after project start · takes 7 days
#19Calibrated JudgmentLowTo Do
The trainable skill of expressing uncertainty so that your stated probabilities and intervals match reality—your 90% confidence bounds should contain the truth 90% of the time.
Assignee: Unassigned · due 167 days after project start · takes 3 days
#20Value of InformationHighTo Do
Use the economic lens for deciding whether a measurement is worth taking at all, and how much precision to buy.
Assignee: Unassigned · due 170 days after project start · takes 14 days
#21Reporting Transparency & ReplicabilityLowTo Do
Documenting your analytic choices and results completely enough that someone else could reproduce them and judge their meaning.
Assignee: Unassigned · due 184 days after project start · takes 3 days
#22Quality of Inference & DecisionsHighTo Do
The ultimate payoff: whether the conclusions and decisions your measurement supports are actually trustworthy and profitable.
Assignee: Unassigned · due 187 days after project start · takes 14 days