Session 9 – The workshop day
The day runs as the one-day course sets it out, in four ninety-minute parts that contain the five tasks, with one difference. At the preregistration moment in Part 4 you return to the preregistration block that you completed and dated in session 8: you read it against the menu once more, re-date it, and then go on to build the model, run it and report the result.
The workshop comes this late in the term because its five tasks ask for decisions, and a term is the one format that can prepare them. By session 9 you already have R and a document to write in, have fitted the estimators that the menu offers and know what each one assumes, have an estimand and a graph, and have a specification declared in the preregistration block of your report. The ninety-minute parts can therefore go on the choosing itself, and because everyone arrives with a specification they can already defend, each dot on the class chart is an analysis with a justification behind it. If the timetable cannot release a whole day, Parts 1 and 2 run in the session-9 slot and Parts 3 and 4 in the session-10 slot, and the closing session takes an eleventh week.
| Work through | What you do | Timing |
|---|---|---|
| The one-day course | Follow its four parts as written. | a full day |
| Slides | The full deck of the course. The optional R tutorial in Part 2 and the methodological tutorial in Part 3 cover what sessions 2 and 4 have already done, so the class will pass over them; the multiverse tutorial in Part 4 stays in, because the module that goes with it comes only in session 10. | throughout |
| Exercise overview, the workspace and the Browser lab | Tasks 1 to 5: get yourself running; meet the panel, name the estimand, draw the graph; reproduce the result recorded by Multi100; work the menu; then preregister one specification, run it and report it. | throughout |
| The Multiverse | Keep the page open from Part 3 onwards. You submit to it twice, and your dots are drawn as an open circle for the exploration and a filled one for the preregistered analysis. | throughout |
By the end of the session you have reproduced the value recorded by Multi100 for analyst C6HJR, t = −3.804 on the 270-country-year panel. You also have an exploratory dot and a declared dot on the class chart, and a report containing the preregistration written in session 8, now with its result underneath it.