Coding Guide
Cognitive Operation |
Examples |
Code |
Core Cognitive Skills |
||
Questioning |
|
Q |
Connecting |
|
C |
Sourcing |
|
S |
Inferencing |
|
I, |
Considering multiple perspectives |
|
MP |
Recognizing limits to knowledge |
|
H |
Other Cognitive Skills |
||
Empathy |
|
Empathy |
Strategizing |
|
St |
Narration |
|
N |
Revision |
|
R |
Causation |
|
CN |
Other Cognitive Operations |
||
Elaboration |
|
E |
Comprehension Monitoring |
|
Comp, |
Text Description |
|
TD |
Uncategorizable/Irrelevant |
|
X |
How we did it:
We (my coding partner and I) came up with the guide in a process that looked like this:
- I did some reading on think alouds.
- I made a preliminary coding guide that identified maybe 8 cognitive processes.
- We read the Think Alouds transcripts and found that students were doing things we didn't have codes for.
- We tentatively labeled these unexpected processes.
- I did some more reading to compare our labels with ones other people used in the literature.
- We made a final coding guide.
The six skills I was teaching in the course and was particularly interested in looking for in the TAs are the first six skills on the guide.
In the first study, my partner was a respected (i.e., well-published) professor from another institution who teaches introductory history courses in traditional ways but is open-minded and interested in teaching innovations. I paid him $500 for a couple days of work. We decided on degrees of proficiency in a process that looked like the following.
- We agreed that “1” would indicate the proficiency of someone who knows almost nothing about historical thinking, who shows no evidence of the skill, or severely misunderstands the skill and how to do it; and “5” that of an average expert historian, i.e., ourselves.
- We read half the transcripts (3) to get a feel for what students were doing, and talked about our impressions.
- We made a rubric to fill in the gaps between 1 and 5.
- We proceeded to read the transcripts, independently ranking the students' performances accoring to our rubric.
Later, when we compared our rankings, if we differed on a ranking, then we talked the ranking through with evidence from the transcripts, until we agreed on a 1–5 ranking for the particular performance in question.
With these decisions made, in a later study to replicate the first one, I used a very sharp student as my partner. I found her to be tougher with rankings than my earlier colleague.