Hi guys, at first I want to apologize for my broken English, here is my problem,
My girlfriend is writing her bachelor thesis in animal science (I am doing my PhD thesis but in different field of science – electrical engineering). She held an experiment where she have tested 17 puppies (total of 3 litter (different breeds)) in order to define character type of each. Test is known as “Volhard’s Puppy Aptitude Testing” (http://www.volhard.com/pages/pat.php). To be short there are 10 different situation and referee should decide how dog react to each, scores vary from 1 to 6. The aim of this work is to find out if character types vary in same litter (brothers and sisters) same as they vary between totally unrelated puppies.
She already calculated some statistical data like averages in litter, standard deviation (std), std in litter (table 1).
To define relationship between puppies’ test result she have calculated the correlation coefficients between same and different puppies (table 2-5).
She has decided to count correlation coef. >0.7 (strong positive dependency) and calculate percentages between puppies in litters and percentages between unrelated puppies.
Some professors (definitely not math) during the consultation asked for value of significance (knows as p-value). I am not sure but as I understand suggested approach would eliminate extremely high and low values (outliers) from data. In this particular test it is important to have all values from 1 to 6.
QUESTION
My question is: should significance approach be applied? Is it beneficial in particular case? If no what answer should be given to professor?
Thant you for your time, any comments, thoughts or suggestions are welcome.
Table 1
Table 2
Table 3
Table 4
Table 5
My girlfriend is writing her bachelor thesis in animal science (I am doing my PhD thesis but in different field of science – electrical engineering). She held an experiment where she have tested 17 puppies (total of 3 litter (different breeds)) in order to define character type of each. Test is known as “Volhard’s Puppy Aptitude Testing” (http://www.volhard.com/pages/pat.php). To be short there are 10 different situation and referee should decide how dog react to each, scores vary from 1 to 6. The aim of this work is to find out if character types vary in same litter (brothers and sisters) same as they vary between totally unrelated puppies.
She already calculated some statistical data like averages in litter, standard deviation (std), std in litter (table 1).
To define relationship between puppies’ test result she have calculated the correlation coefficients between same and different puppies (table 2-5).
She has decided to count correlation coef. >0.7 (strong positive dependency) and calculate percentages between puppies in litters and percentages between unrelated puppies.
Some professors (definitely not math) during the consultation asked for value of significance (knows as p-value). I am not sure but as I understand suggested approach would eliminate extremely high and low values (outliers) from data. In this particular test it is important to have all values from 1 to 6.
QUESTION
My question is: should significance approach be applied? Is it beneficial in particular case? If no what answer should be given to professor?
Thant you for your time, any comments, thoughts or suggestions are welcome.
Table 1
Test | Statistical analysis | ||||||||||||
Dog marking | situation 1 | situation 2 | situation 3 | situation 4 | situation 5 | situation 6 | situation 7 | situation 8 | situation 9 | situation 10 | Avg. In litter | Standard deviation | Standard deviation in litter |
z1 | 1 | 1 | 2 | 3 | 2 | 4 | 1 | 2 | 3 | 4 | 3,04 | 1,10 | 1,04 |
z2 | 3 | 3 | 2 | 3 | 3 | 3 | 5 | 3 | 2 | 3 | 0,77 | ||
z3 | 2 | 3 | 6 | 3 | 5 | 3 | 6 | 3 | 3 | 6 | 1,48 | ||
z4 | 3 | 3 | 3 | 3 | 3 | 3 | 1 | 2 | 5 | 4 | 1,00 | ||
z5 | 2 | 2 | 2 | 3 | 2 | 4 | 3 | 3 | 4 | 4 | 0,83 | ||
b1 | 1 | 1 | 2 | 2 | 2 | 2 | 6 | 2 | 1 | 2 | 2,43 | 1,37 | 1,28 |
b2 | 2 | 1 | 2 | 2 | 3 | 4 | 3 | 5 | 3 | 6 | 1,45 | ||
b3 | 1 | 1 | 2 | 2 | 2 | 3 | 1 | 2 | 3 | 4 | 0,94 | ||
b4 | 1 | 1 | 4 | 2 | 3 | 1 | 1 | 3 | 3 | 5 | 1,36 | ||
a1 | 4 | 6 | 4 | 4 | 4 | 6 | 1 | 4 | 4 | 3 | 4,24 | 1,34 | 1,26 |
a2 | 6 | 6 | 3 | 4 | 4 | 6 | 6 | 4 | 4 | 4 | 1,10 | ||
a3 | 5 | 6 | 2 | 4 | 4 | 4 | 6 | 5 | 4 | 3 | 1,19 | ||
a4 | 4 | 5 | 4 | 6 | 4 | 1 | 6 | 5 | 3 | 5 | 1,42 | ||
a5 | 2 | 6 | 3 | 4 | 4 | 3 | 6 | 2 | 2 | 2 | 1,50 | ||
a6 | 2 | 2 | 3 | 3 | 4 | 4 | 6 | 6 | 3 | 2 | 1,43 | ||
a7 | 5 | 6 | 4 | 4 | 4 | 5 | 6 | 5 | 6 | 6 | 0,83 | ||
a8 | 5 | 6 | 5 | 6 | 4 | 2 | 6 | 5 | 3 | 4 | 1,28 |
Table 2
Dog marking | z1 | z2 | z3 | z4 | z5 |
z1 | 1,00 | ||||
z2 | -0,35 | 1,00 | |||
z3 | 0,06 | 0,26 | 1,00 | ||
z4 | 0,55 | -0,77 | -0,20 | 1,00 | |
z5 | 0,80 | 0,00 | 0,00 | 0,36 | 1,00 |
Table 3
Dog marking | b1 | b2 | b3 | b4 |
b1 | 1,00 | |||
b2 | 0,15 | 1,00 | ||
b3 | -0,24 | 0,73 | 1,00 | |
b4 | -0,18 | 0,54 | 0,67 | 1,00 |
Table 4
Dog marking | a1 | a2 | a3 | a4 | a5 | a6 | a7 | a8 |
a1 | 1,00 | |||||||
a2 | 0,14 | 1,00 | ||||||
a3 | -0,06 | 0,76 | 1,00 | |||||
a4 | -0,58 | -0,13 | 0,30 | 1,00 | ||||
a5 | -0,10 | 0,44 | 0,55 | 0,41 | 1,00 | |||
a6 | -0,42 | 0,03 | 0,32 | 0,12 | 0,19 | 1,00 | ||
a7 | -0,18 | 0,47 | 0,48 | 0,06 | 0,13 | -0,04 | 1,00 | |
a8 | -0,35 | 0,06 | 0,41 | 0,89 | 0,55 | 0,05 | -0,06 | 1,00 |
Table 5
Dog marking | z1 | z2 | z3 | z4 | z5 | b1 | b2 | b3 | b4 |
b1 | -0,22 | 0,85 | 0,59 | -0,73 | 0,10 | ||||
b2 | 0,61 | 0,09 | 0,28 | 0,07 | 0,67 | ||||
b3 | 0,93 | -0,41 | 0,21 | 0,64 | 0,78 | ||||
b4 | 0,46 | -0,48 | 0,55 | 0,44 | 0,21 | ||||
a1 | 0,20 | -0,58 | -0,60 | 0,37 | -0,09 | -0,76 | -0,26 | 0,08 | -0,22 |
a2 | -0,34 | 0,59 | -0,37 | -0,36 | -0,03 | 0,22 | -0,23 | -0,45 | -0,86 |
a3 | -0,60 | 0,65 | -0,40 | -0,51 | -0,17 | 0,29 | -0,25 | -0,65 | -0,76 |
a4 | -0,44 | 0,46 | 0,29 | -0,42 | -0,31 | 0,40 | -0,11 | -0,40 | 0,09 |
a5 | -0,50 | 0,60 | 0,23 | -0,53 | -0,37 | 0,52 | -0,53 | -0,59 | -0,52 |
a6 | -0,16 | 0,54 | 0,19 | -0,70 | 0,13 | 0,69 | 0,31 | -0,19 | -0,15 |
a7 | -0,03 | 0,31 | 0,00 | 0,12 | 0,45 | 0,17 | 0,24 | 0,11 | -0,12 |
a8 | -0,70 | 0,40 | 0,11 | -0,55 | -0,60 | 0,31 | -0,46 | -0,71 | -0,20 |