Sep 28, 2016 17:45
7 yrs ago
2 viewers *
Spanish term
Spain
cruce
Spanish to English
Tech/Engineering
Mathematics & Statistics
Statistical analysis
This is a justification of why the GLM was not applied to analyze the data collected in the study. "Cruce" appears twice, but in both cases, the obvious correspondence "cross", sounds a bit strange to me (but I could be wrong). Any help would be appreciated. Here is the context
Tenemos que insistir en que aplicar un GLM a los datos que se están estudiando, no es viable (lo que no quiere decir que para otros estudios retrospectivos con un número mayor de casos, sí podría ser adecuado). La razón es muy simple: en el estudio se consideran 5 factores con al menos dos niveles en cada factor. Considerar todas las interacciones posibles, llevaría a considerar al menos 2 x 2 x 2 x 2 x 2 ***cruces***, es decir, 32 posibles interacciones entre los factores cruzados. Dado que en número de casos estudiados es de 524, en el caso más favorable de que estos estuvieran repartidos como en un diseño experimental planificado, habría un número menor de 20 casos por ***cruce***.
Tenemos que insistir en que aplicar un GLM a los datos que se están estudiando, no es viable (lo que no quiere decir que para otros estudios retrospectivos con un número mayor de casos, sí podría ser adecuado). La razón es muy simple: en el estudio se consideran 5 factores con al menos dos niveles en cada factor. Considerar todas las interacciones posibles, llevaría a considerar al menos 2 x 2 x 2 x 2 x 2 ***cruces***, es decir, 32 posibles interacciones entre los factores cruzados. Dado que en número de casos estudiados es de 524, en el caso más favorable de que estos estuvieran repartidos como en un diseño experimental planificado, habría un número menor de 20 casos por ***cruce***.
Proposed translations
(English)
| 4 -1 | crossover |
neilmac
|
| 4 +1 | combination |
Francois Boye
|
| 4 | two-factor interaction term |
DLyons
|
Proposed translations
-1
47 mins
Selected
crossover
Used to describe interactions:
http://www.theanalysisfactor.com/interactions-main-effects-n...
--------------------------------------------------
Note added at 48 mins (2016-09-28 18:34:31 GMT)
--------------------------------------------------
NB: DLyons is the expert in this area (soy de letras)...
http://www.theanalysisfactor.com/interactions-main-effects-n...
--------------------------------------------------
Note added at 48 mins (2016-09-28 18:34:31 GMT)
--------------------------------------------------
NB: DLyons is the expert in this area (soy de letras)...
Note from asker:
| Thank you very much for your help. |
Peer comment(s):
| disagree |
DLyons
: Sorry to disagree Neil - but this this has a specific meaning which is not what the ST is saying. See above.
2 days 12 hrs
|
4 KudoZ points awarded for this answer.
Comment: "I checked with the client, and this is the term that he prefers."
43 mins
two-factor interaction term
This seems like a complete 2^5 factorial design.
+1
13 hrs
Francois Boye
United States
United States
Specializes in field
United States
An economic geek with social skills
Native in: French 
Works in: English to French, French to English, Spanish to English, and 1 more.
Related KudoZ points
:
72
combination
https://en.wikipedia.org/wiki/Combination
GLM = General Linear Model
cruce = the combination of more than two variables to run a GLM regression
--------------------------------------------------
Note added at 13 hrs (2016-09-29 06:57:08 GMT)
--------------------------------------------------
https://en.wikipedia.org/wiki/General_linear_model
--------------------------------------------------
Note added at 1 day13 hrs (2016-09-30 07:27:40 GMT)
--------------------------------------------------
Erratum:'COMBINATION OF TWO DIFFERENT VARIABLES TO RUN A GLM REGRESSION' instead of 'the combination of more than two variables to run a GLM regression'
--------------------------------------------------
Note added at 1 day13 hrs (2016-09-30 07:39:59 GMT)
--------------------------------------------------
To compute the number of combinations, on can build a matrix with 10 columns and rows. Each non-diagonal element of the matrix is a combination of two different variables. So the number of combinations is equal to 10X10 -10 = 100-10 = 90. 100 is the number of combinations (cruce) and there are 10 diagonal combinations of IDENTICAL variables
GLM = General Linear Model
cruce = the combination of more than two variables to run a GLM regression
--------------------------------------------------
Note added at 13 hrs (2016-09-29 06:57:08 GMT)
--------------------------------------------------
https://en.wikipedia.org/wiki/General_linear_model
--------------------------------------------------
Note added at 1 day13 hrs (2016-09-30 07:27:40 GMT)
--------------------------------------------------
Erratum:'COMBINATION OF TWO DIFFERENT VARIABLES TO RUN A GLM REGRESSION' instead of 'the combination of more than two variables to run a GLM regression'
--------------------------------------------------
Note added at 1 day13 hrs (2016-09-30 07:39:59 GMT)
--------------------------------------------------
To compute the number of combinations, on can build a matrix with 10 columns and rows. Each non-diagonal element of the matrix is a combination of two different variables. So the number of combinations is equal to 10X10 -10 = 100-10 = 90. 100 is the number of combinations (cruce) and there are 10 diagonal combinations of IDENTICAL variables
Peer comment(s):
| agree |
DLyons
: The client seems to want a relatively non-technical term.
2 days 23 mins
|
Discussion