Wilcox test HELP!!!

okay so I am doing psychology and I have been doing the stroop experiment, you know say words in different colors and see how long it take anyway I am now going to calculate the results and I am suppose to do it with the wilcox test but I can't figure it out so

if I have for ex
1st times 2nd times difference Rank
14,583 26,803 12,22 1
11,296 24,305 13,009 2
13,413 27,26 13,857 3
and so on

I will calculate raking later so I just filled in random nr, so next step calculate T is what gets me
T= the sum of the ranks of the less frequent sign :confused:

which I guess is + or - but all my differences are plus, the participants were always slower on the second timed test so how am I suppose to add the nr up :confused:

Please help anyway you can I am terrible at these things, maths :s-smilie:

. So before you tell me to, I don't have a teacher to ask or anyone else. I am studying by myself.

Reply 1

wellpastmybedtime

I am not quite sure I understand you.

You should have two sets of data from a repeated measures design experiment ( ie one group of participants and they did two tasks?) Just checking.

So, list the scores for each condition.

Calculate the difference between each pair of scores.

Rank the differences from smallest to largest ignoring the +/- signs.

Then add together all the ranks given a +.

then add together all the ranks given a minus.

you will then have a total of minus ranks ( ones with minus signs ) and a total for plus ranks. ( ones with plus signs)

Take the smaller of the two totals. Count the number of pairs ( not counting ties ( that is any where both conditions got the same scores)

Find a table book of critical values for the wilcoxon - compare your smallest number to that given in critical values at whatever significance level you are setting ( p : =/+ 0.05 usually)

Reply 2

sheshbesh

wellpastmybedtime

I am not quite sure I understand you.

You should have two sets of data from a repeated measures design experiment ( ie one group of participants and they did two tasks?) Just checking.

So, list the scores for each condition.

Calculate the difference between each pair of scores.

Rank the differences from smallest to largest ignoring the +/- signs.

Then add together all the ranks given a +.

then add together all the ranks given a minus.

you will then have a total of minus ranks ( ones with minus signs ) and a total for plus ranks. ( ones with plus signs)

Take the smaller of the two totals. Count the number of pairs ( not counting ties ( that is any where both conditions got the same scores)

Find a table book of critical values for the wilcoxon - compare your smallest number to that given in critical values at whatever significance level you are setting ( p : =/+ 0.05 usually)

Life saver, but when you say pair of scores you mean Participant Ones colour score minus Participant Ones noun score? I'm THAT lost.

Also, if you get any zero results do you exclude that participant? Sorry about all the questions. :o:

Reply 3

ghy56_

OP

I know it is hard to explain but I ONLY have +.
so
condition 1 14,583 seconds
Condition 2 26,803 seconds
Difference 12,22 seconds

all my differences are something similar 13,009, 14,456 and so on. Aren't they all considered +, to be positive?

so I rank the differences but then when it comes to adding all the ranks with + and all with - together it doesn't work because they are all +. I don't know maybe I am just thinking completely wrong but I can't seem to figure out where to go from here hope you understand better now, thank you

Reply 4

wellpastmybedtime

Life saver, but when you say pair of scores you mean Participant Ones colour score minus Participant Ones noun score? I'm THAT lost.

Also, if you get any zero results do you exclude that participant? Sorry about all the questions.

Basically yes to both questions.

You used the same participants in two different conditions - the Colour test and the noun test - so for each subject ( sorry participant - you can tell how long I have been doing psychology from that!) you will have two scores.

You take second score from the first.

Where you have a 0 difference in the pairs after subtracting you discount that pair when calculating the number for the critical values tables.

Hope that is clear.

Reply 5

ghy56_

OP

wellpastmybedtime

I am not quite sure I understand you.

You should have two sets of data from a repeated measures design experiment ( ie one group of participants and they did two tasks?) Just checking.

So, list the scores for each condition.

Calculate the difference between each pair of scores.

Rank the differences from smallest to largest ignoring the +/- signs.

Then add together all the ranks given a +.

then add together all the ranks given a minus.

you will then have a total of minus ranks ( ones with minus signs ) and a total for plus ranks. ( ones with plus signs)

Take the smaller of the two totals. Count the number of pairs ( not counting ties ( that is any where both conditions got the same scores)

Find a table book of critical values for the wilcoxon - compare your smallest number to that given in critical values at whatever significance level you are setting ( p : =/+ 0.05 usually)

I tried to explain abit more in my previous post but looking at it now I think I mean that all of my differences are - minus, it doesn't solve my problem about how I am so suppose to add all + together and all - together but maybe it makes sense more for someone who knows

Reply 6

wellpastmybedtime

OK , so all of your differences are minus - so in each case the subjects did better in one condition than the other?

Thats not a problem.

Rank them. Add them all up . Count N as the total number of participants ( I take it there were no tied scores either counting a difference of 0?) . Use that as your figure to compare to critical values.

Reply 7

ghy56_

OP

All the participants took longer on the second test so the difference is always negativ, which means I have no positive to add up, usch I can't explain it any more I hope someone understands what I mean and can tell me what I am doing wrong

Reply 8

ghy56_

OP

Thank YOU!!!!!!!!

Reply 9

wellpastmybedtime

your plus rank will be zero.

You should have a significant result by the way.

I am taking it the scores went in the direction of your hypothesis - or did you use a two tailed hypothesis ( in which case it wont matter) because you will be simply stating a difference and difference you certainly would seem to have.

Reply 10

ghy56_

OP

yes I does fit well with my hypothesis it's one tailed by the way anyway thsnk you again and there is some rep coming your way

Wilcox test HELP!!!

Related discussions

Latest

Trending

Trending

Articles for you