24 |
- perhaps 16 texts, each read twice by the subject with |
- perhaps 16 texts, each read twice by the subject with |
25 |
randomly assigned conditions? |
randomly assigned conditions? |
26 |
- 48 trials could also be just possible |
- 48 trials could also be just possible |
27 |
|
- how many texts do we need? |
28 |
|
- there's a new version with more texts |
29 |
|
(and a typo fixed) in |
30 |
|
http://himalia.it.jyu.fi/~jvkujala/texts/ |
31 |
|
- there are 25 texts, but perhaps we should only |
32 |
|
use 18 or less of most similar lengths |
33 |
|
(see showlen Makefile target) |
34 |
- Think about randomization |
- Think about randomization |
35 |
- can we have a subject read the same text twice in a row? |
- can we have a subject read the same text twice in a row? |
36 |
- do we need to enforce the same number of trials for each |
- do we need to enforce the same number of trials for each |
38 |
twice by a subject? |
twice by a subject? |
39 |
Or can we trust just on the number of trials, to be |
Or can we trust just on the number of trials, to be |
40 |
approximately non-skewed? |
approximately non-skewed? |
41 |
- perhaps should first read each of the 16 different texts |
- perhaps a subject should first read each of the |
42 |
and then again, in the same order, (but with different |
16 different texts and then again, in the same order, |
43 |
permutation of conditions). That way the time between |
(but with a different permutation of conditions). |
44 |
readings is the same for each text. |
That way the time between readings is the same for each text. |
45 |
- Create a program that reads lines of the form |
- Create a program that reads lines of the form |
46 |
<filename> <min-time> <max-time> <list-of-keys> |
<filename> <min-time> <max-time> <list-of-keys> |
47 |
and for each line |
and for each line |