karusmle
Forum Guru
Topics: 296
Posts: 1,001
|
hi friends
i m getting confussed with those statistical tests questions ,can anybody explain to me the statistical tests in a simpler way,(i read kaplan but still my concepts r not clear with this)
paired t test ,chi square,student paired t test ,ANOVA,meta analysis , basic meaning for nominals ,etc etc
while doing Q book , i got all the qusetions with statistical tests wrong 
|
| demnat
Forum Junior
Topics: 6
Posts: 35
|
Same problem with me! Got 30% on qbank on statistic!!!
would be great if somebody can explain!
|
| drsnq
Forum Elite
Topics: 24
Posts: 168
|
I also face the same problem...
___________________
drsnq
|
| msyamp
Forum Fanatic
Topics: 60
Posts: 1,462
|
hi guys i am taking my stats course now in my MPH. i will be finishing all these by next week.I think i get an understanding on where each test is used.hope I can clarify you then,.
___________________
If you think you can You can! If you think you cant you are right again!!
|
| reet
Forum Fanatic
Topics: 152
Posts: 1,411
|
msyamp,,,
i will wait for ur post..
thanks
___________________
Live as u were to die tommorow. Learn as if u were to live forever.
|
| msyamp
Forum Fanatic
Topics: 60
Posts: 1,462
|
here i am!!! let us think we have two classes one of india and one on america.if you administer a half yearly and then a final exam to the india class.If you want to compare the means of the exams of these two tests we use a paired t test.that is paired t test is used to compare means of pre and post tests of the same group.if you do it for students it is called student paired t test.
chi square is done for a 2 * 2 table.for example you have exposed and not exposed, diseased and non diseased as in case control. or in cohort. the thing here is not the kind of study but you have to have a 2 * 2 table and to have such a table you should be using nominal level data.
ANOVA analysis of varience is done in case where you have two or more groups and you are comparing the means scores of all these groups. here you can determine whether the means of all the groups are equal or not equal.but cannot identify which group is different from other groups.
to determine differences between individual groups you should do a t test between those groups. now because you are doing it between different groups it is an unpaired t test.
meta analysis is the analysis of data of similar kind of studies over a period of time and done at different places. you mix the results of all the studies on the same topic and come to a conclusion. anlysis of all studies combined.
nominal is name.that is you cannot put them in order.for example syam, ram ,prem, john.
i hope you get to know all other kind of data. ordinal interval and ratio.
___________________
If you think you can You can! If you think you cant you are right again!!
|
| karusmle
Forum Guru
Topics: 296
Posts: 1,001
|
thanks alot , i got these now  into my head .............
|
| reet
Forum Fanatic
Topics: 152
Posts: 1,411
|
msyamp thanks...
___________________
Live as u were to die tommorow. Learn as if u were to live forever.
|
|
| |
| | | | | | | | |
