Data to Input in PostableConverter Spreadsheet

[Katalina Fernandez McInerney]

Hi,

Thank you for hosting this group. 

I am using Set E and the data I am getting does not seem to match the data asked for in the converter spreadsheet. Here is an example of the summary data I am seeing at the bottom of each file:

Lure Bin Statistics (raw counts)
L1O L1S L1N L2O L2S L2N L3O L3S L3N L4O L4S L4N L5O L5S L5N LNR
5        1      1      7     0     1      1      6     1      4     4      0     0      7     1
L1-NR L2-NR L3-NR L4-NR L5-NR
1            0           0         0         0

Response totals (out of 64 expected)

Target: 40
Lure: 39
Foil: 39


Percent correct (corrected): 62.71

Percent correct (raw): 61.67
Bias metric (S|L-S|F): 25.64

Is any of this data what you would input into the Bins, Foil, LDIs, or Recognition columns?

Overall, I am having a difficult time understanding the MST data output. Is there a good reference for this out there?

Thanks again.

[Craig E.L. Stark]

OK, so let's break that down for you...  That table there is showing how people responded to the various lure bins and there should be a bit more in the file that you've not attached (and that may make your life easier), but we can do the decoding (BTW, what version of the MST are you running? Some have nicer / richer output at the end).

L1O = Lure bin 1, subject called it "old"

L1S = Lure bin 1, subject called it "similar"

...

L4N = Lure bin 4, subject called it "new"

So, on the "bin based conversion" portion of the spreadsheet, it's asking for you to enter in the p(Sim|Lure) data.  That is the probability of saing "similar" given that it's a lure.  Let's take your data here we see that for Lure bin 1 (L1x) the subject said "old" 5 times, "similar" once, and "new" once.  So P(Sim|Lure) if you need to compute it manually would be 1/7.  The next bin would be 0/7 and bin 3 would be 6/8, etc.

The LDI metric = "Bias metric" = p(Similar|Lure) - p(Similar|Foil) = 25.64 above and my guess is, this should really be 0.2564 when entered into the spreadsheet.  The "Percent correct" is your REC (recognition score).

Craig