Can members keep their own local questions?

An independent member organization may already run its own surveys and report to its own board. A chapter may need questions suited to its local events. Combining evidence does not require every team to abandon those questions.

Agree on a small shared core: organization or group ID, reporting period, measure definition and the status of the evidence. Keep local fields alongside it. Map an existing question only when its meaning actually matches; a five-point satisfaction score cannot silently become a measure of value for dues.

Use the data-dictionary guide to record definitions, owners and missing-value rules. For this pilot, complete four fields rather than the whole reference. In the network practice file, “unique participants” and “attendance occasions” need separate definitions. N03’s 12 attendance occasions cannot be added to N01’s 40 and N02’s 30 people.

Which groups can you compare from your AMS fields?

In short: compare groups your AMS already names, and let the list tell you how many members were eligible.

Chapter and member type came in with your member list in lesson 2, and every survey answer sits on the same member record from lesson 4. A comparison by chapter then uses one definition of the group for every member. The AMS count for each group is the eligible population; the members who gave a usable answer are the response base. A benchmark shows both.

A second wave to the same members works the same way. The question, the period and the group stay fixed, so a change between waves compares the same people instead of two spreadsheets to reconcile.

How do you compare groups that answered in different numbers?

This is an optional extension. If your goal includes a benchmark, use this additional exercise. The A/B/C figures are a separate synthetic dataset, not a calculation from the six-person or three-organization practice files. All groups used the same member-value question and period. Their coverage still differs.

This exercise withholds a group result when fewer than ten usable answers are available. That is an illustrative rule, not a universal privacy standard. Small groups, distinctive quotes and overlapping reports can still disclose information. The actual release policy needs to fit the data and the audience.

A and B together have 30 positive answers among 45 usable answers: 66.7%. Do not average 60% and 80% to produce 70%. Describe the restricted A+B population: 45 usable answers from 70 eligible members. The complete exercise includes 90 eligible members across all three groups. Avoid releasing another total that lets a reader reconstruct a withheld small-group result.

The cohort comparison exercise shows how respondent mix can produce an apparent overall difference. Use it before interpreting A versus B. Bring back the eligible population, response base, period and one possible alternative explanation.

What should the report give your reader?

For the individual-member path, report which members meet the question’s conditions, the supporting records and the permitted next action. For the member-organization path, return the organization’s reviewed evidence, unresolved questions and useful next steps. The report below illustrates the optional benchmark extension; you do not need it to finish either core path.

For a network of organizations, return that organization’s reviewed submission, its open questions and an approved comparison. Let it correct its own evidence and review a local report built from it.

In the network exercise, report 70 participants as reported by two of three organizations, with cross-organization overlap unknown. Return N03 a clarification request and its own current summary. Do not turn incomplete coverage into a league table.

Could the report become a paid member benefit?

A benchmark, salary study or industry report may support non-dues revenue when it helps a reader make a recurring decision. Producing a chart does not establish that demand. Begin with a report prototype and ask potential readers what decision it changes, what comparison is missing and how they would use it.

Then test willingness to pay for a clearly described offer, update frequency and delivery format. Include the cost of collection, review and support. Commercially sensitive comparisons, including competitor salary or pricing information, need an appropriate legal and disclosure review before collection or release. This lesson’s member-value example does not settle those requirements.