Question 48·Hard·Command of Evidence
Researchers investigating whether text-message reminders affect attendance at community financial-literacy workshops classified registrants as either first-time or returning participants. Within each subgroup, the researchers randomly assigned equal numbers of registrants to receive a reminder or no reminder.
Before collecting the data, the researchers predicted that reminders would increase attendance by about the same number of percentage points in both subgroups. Therefore, they expected the attendance-rate gap between first-time and returning registrants to be about the same with or without reminders.
| Registrant subgroup | Reminder: registered | Reminder: attended | No reminder: registered | No reminder: attended |
|---|---|---|---|---|
| First-time | 160 | 112 | 160 | 88 |
| Returning | 140 | 112 | 140 | 105 |
Which choice best describes information from the table that weakens the researchers’ prediction?
For quantitative command-of-evidence questions, translate counts into the rates or proportions named in the claim before evaluating the choices. When a prediction has multiple parts, check each part: compare the treatment effect within relevant groups, and then determine whether any stated gap or pattern changes across conditions.
Hints
Break the prediction into parts
Determine what the prediction says about both the effect of reminders within each subgroup and the gap between the subgroups.
Use rates rather than attendee counts
Divide the number who attended by the number registered in each table cell. Because the subgroup sizes differ, equal attendee counts do not imply equal attendance rates.
Compare changes and gaps
First compare each subgroup’s reminder and no-reminder rates. Then compare the difference between subgroup rates under the two conditions.
Step-by-step Explanation
Identify the comparisons required by the prediction
The prediction concerns attendance rates, not raw numbers of attendees. It also makes two linked claims: the subgroups will experience similar-sized gains, and the gap between their attendance rates will remain about the same.
Calculate the subgroup attendance rates
For first-time registrants, the attendance rates were with reminders and without reminders.
For returning registrants, the rates were with reminders and without reminders.
Compare the gains from reminders
The first-time subgroup’s attendance rate increased by percentage points. The returning subgroup’s rate increased by only percentage points, so the gains were not about the same.
Compare the attendance-rate gaps
Without reminders, the gap was percentage points. With reminders, it was percentage points. Because the gap decreased substantially, the table weakens the prediction that it would remain about the same.