[FieldTrip] Difference in trial number per condition

[Zsolt Turi]

Hi Stan and Diego,

thank you for the input(s), I will try out these procedures.

Cheers,
Zsolt

2016-06-17 11:02 GMT+02:00 Diego Lozano-Soldevilla <
dlozanosoldevilla at gmail.com>:

> Hi Zsolt,
> To complement Stan's suggestions, in case of power differences between
> conditions, you could use ft_stratify:
> http://www.fieldtriptoolbox.org/example/stratify
> best,
> Diego
>
>
>
> On 17 June 2016 at 10:51, Pelt, S. van (Stan) <stan.vanpelt at donders.ru.nl>
> wrote:
>
>> Hi Zsolt,
>>
>>
>>
>> The amount to which this may be a concern will depend on the measures you
>> want to compare. Also, the risk of drawing incorrect (qualitative)
>> conclusions will of course mainly be large around your statistical
>> threshold value (e.g. p=0.05). Power differences might become biased if
>> groups have unequal size, power begin a squared value. See for a suggested
>> numerical correction method for this, when comparing power spectra or e.g.
>> Bokil et al., 2007 (http://www.ncbi.nlm.nih.gov/pubmed/16945422), but I
>> assume there are also other solutions going around.
>>
>>
>>
>> Best,
>>
>> Stan
>>
>>
>>
>> *From:* fieldtrip-bounces at science.ru.nl [mailto:
>> fieldtrip-bounces at science.ru.nl] *On Behalf Of *Zsolt Turi
>> *Sent:* vrijdag 17 juni 2016 9:32
>> *To:* FieldTrip discussion list
>> *Subject:* [FieldTrip] Difference in trial number per condition
>>
>>
>>
>> Hi,
>>
>>
>>
>> I was wondering if FieldTrip has an optimized solution for such
>> situations when the amount of trials in one condition markedly differs from
>> the number of trials in the other (e.g., the ratio is 1:3 or 1:4).
>>
>>
>>
>> On the website I have found that this issue was already discussed,
>> however, it is not clear to me whether this was indeed addressed by a
>> FieldTrip function. Below is the link about the question I was referring
>> to:
>>
>>
>>
>>
>> http://www.fieldtriptoolbox.org/faq/how_does_a_difference_in_trial_numbers_per_condition_affect_my_statistical_test
>>
>>
>>
>> In case there is no such a function, could someone provide me a rough
>> procedure that is methodologically acceptable in the scientific community
>> to compensate for this issue?
>>
>>
>>
>> I am interested both in conceptual and implementational aspects. I would
>> like to compare TFR or WPLI measures between two conditions and use
>> non-parametric cluster-based permutation test for statistical analysis. My
>> data is characterized by a within-subjects design, in case this is relevant
>> information.
>>
>>
>>
>> Thanks for your input in advance,
>>
>> Zsolt
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> https://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>



-- 

Post-doctoral Research Fellow
Department of Clinical Neurophysiology
Georg-August University, Göttingen
Robert-Koch-Str. 40
37075 Goettingen
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20160617/1ef14f9b/attachment-0002.html>