Login or register (free and only takes a few minutes) to participate in this question.You will also have access to many other tools and opportunities designed for those who have languagerelated jobs (or are passionate about them). Participation is free and the site has a strict confidentiality policy.
English to English translations [PRO] Science / statistical analysis 

English term or phrase: multisample comparison  (I'd appreciate an explanation and any synonyms or at least a related terms as well, if possible.)
The text is about algal growth inhibition test (a toxicity test).
"The No Observed Effect Concentration is determined by a suitable statistical procedure for multisample comparison (e.g. analysis of variance and Dunnett's test), using the individual replicates values of the areas under the growth curves A (see point 2.1) or the specific growth rates u (see point 2.2)." 
 Jörgen SletKudoZ activityQuestions: 132 ( 3 open) ( 30 without valid answers) ( 6 closed without grading) Answers: 97 Estonia
  Local time: 14:48

 Multisample comparisons  Explanation: are statisical methods that compare the numerical results from three or more groups/samples(in the sense of collections of results from a group,in this case)....
...to establish whether or not they are significantly different or not (within a certain degree of probability). What is actually compared is one variable/characteristic (or more than one of the variables) in each group, compared with the corresponding variables in the other groups.
An example of the mean value for a variable in a group would be the mean number of eggs in the nest of a group/sample of nesting birds  another variable might be the mean number of years the mother birds have nested.
Anova and Analysis of Variance are the same thing, although there are various sorts of Anova / Analysis of Variance tests to suit different situations
Anova is particularly used to compare mean values for a variable (sometimes more than one) between three or more groups/samples
(as in:
http://www.anselm.edu/homepage/jpitocch/anova/anovapage2.htm...
"ANOVA and Multiple Comparisons
Analysis of Variance  ANOVA
Developed by Fisher while studying agriculture and crop output with different fertlizer treatments, needed test that could evaluate differences between three or more means, developed the F distribution
Why [are there] problems with applying the Student's t test to more than 2 sample/population means....?"
It's normal to use other tests ("t" or "z" depending on size) to compare variables from a pair of groups/samples  anova is used as well for this in some circumstances (as well as for 3 or more groups), but anyway this isn't what is meant by multisample analysis (multisample comparison, etc).
'multiple comparison tests', 'multiple sample comparison tests', and 'multisample comparison [tests]' are names which are sometimes used interchangeably.
'One way Anova' is an Anova test comparing the mean values for one variable between the groups/samples, to see if they are significantly different or not (it compares the groups 'in one way' or aspect).
'Two way Anova' compares mean values for two or more variables between the groups, and so on.
Sometimes an Anova test is made first to establish if there is a significant difference between results for the three or more groups, and this is then followed by a test such as Dunnett to provide additional information (eg: when an Anova test is used to tell you that samples do differ from one another, but doesn't tell you exactly which results are the ones which are different, and more information about this is wanted).
(as in)
http://www.texasoft.com/manual56.htm
"Multiple Comparisons Analysis
When you perform an analysis of variance test, the Ftest (main effects) usually is testing for the presence of a difference between factor means. However, if the test is significant, you do not necessarily know which of the several means can be considered significantly different. Some people have tended to perform multiple ttests to examine the differences between a number of means; however, this approach is incorrect since the pvalues associated with multiple tests are no longer appropriate. Multiple comparison tests are designed to allow you to perform differences between all possible pairs of means using an appropriate and controlled significance level.
WINKS allows you to perform four different kinds of multiple comparison tests. They are:
NewmanKeuls (Newman, 1939, Keuls, 1952)
Tukey (Tukey, 1953)
Scheffé (Scheffé, 1953; 1959)
Dunnett (Dunnett, 1955)"
Dunnett (Dunnett, 1955)
Dunnett's test is a specialized multiple comparison test that allows you to compare a single control group to all other groups. [others eg: Tukey (Tukey, 1953), Scheffé (Scheffé, 1953; 1959) ]
The test you use depends, in part, on your discipline. Some areas of research prefer one multiple comparison test over another. You should consult your literature to see which of these tests is most often used.
http://www.texasoft.com/manual56.htm
 Note added at 20 hrs 32 mins (20031109 14:00:53 GMT) 
typo.error:
 when I said: \"What is actually compared is one variable/characteristic (or more than one of the variables) in each group, compared with the corresponding variables in the other groups.\"
I meant: \" What is actually compared is the value for one variable/characteristic (or more than one of the variables) in each group, compared with the corresponding values in the other groups.\"
 Note added at 21 hrs 4 mins (20031109 14:32:58 GMT) 
Why do we need Anova (and Multisample comparison) ?
There are methods of comparing the mean values of two samples. Sometimes however biologists wish to compare the means of more than two samples. Suppose, for example, we have the length measurements from samples of three races of a species each of which lives on different islands A, B and C. It is possible to compare mean lengths [to see if there are significant differences] by ztests or,if the samples are small, by ttests. We would need to perform the test three times to compare AB, AC and BC. With the help of a calculator the task is not too daunting. Let us imagine instead that we wish to compare the means of seven samples. In this event no less than 21 ztests are required to compare all possible pairs of means. Even if the analyst has sufficient patience to work through this cumbersome treatment, there is an underlying statistical objection to doing so.
For example, if the \"P\"=0.05 (probability level is set at 5%) level of significance, a wrong conclusion will be drawn on average every 20 tests. In 21 ztests there is a good chance of at least one false conclusion. Anova overcomes these problems by allowing comparisons to be made between any number of sample mean values, all in a single test.
(With thanks to \"Practical Statistics For Field Biology\" Jim Fowler, Lou Cohen & Phil Jarvis  pub. Wiley 1998
Chapter 17 Analysis of Variance  Anova) 
 Selected response from: DGK TI United Kingdom Local time: 12:48
 Grading comment Most thorough explanation, thank you ! 4 KudoZ points were awarded for this answer 
 
Discussion entries: 0 

Automatic update in 00:

24 mins confidence:
20 hrs confidence: Multisample comparisons
Explanation: are statisical methods that compare the numerical results from three or more groups/samples(in the sense of collections of results from a group,in this case)....
...to establish whether or not they are significantly different or not (within a certain degree of probability). What is actually compared is one variable/characteristic (or more than one of the variables) in each group, compared with the corresponding variables in the other groups.
An example of the mean value for a variable in a group would be the mean number of eggs in the nest of a group/sample of nesting birds  another variable might be the mean number of years the mother birds have nested.
Anova and Analysis of Variance are the same thing, although there are various sorts of Anova / Analysis of Variance tests to suit different situations
Anova is particularly used to compare mean values for a variable (sometimes more than one) between three or more groups/samples
(as in:
http://www.anselm.edu/homepage/jpitocch/anova/anovapage2.htm...
"ANOVA and Multiple Comparisons
Analysis of Variance  ANOVA
Developed by Fisher while studying agriculture and crop output with different fertlizer treatments, needed test that could evaluate differences between three or more means, developed the F distribution
Why [are there] problems with applying the Student's t test to more than 2 sample/population means....?"
It's normal to use other tests ("t" or "z" depending on size) to compare variables from a pair of groups/samples  anova is used as well for this in some circumstances (as well as for 3 or more groups), but anyway this isn't what is meant by multisample analysis (multisample comparison, etc).
'multiple comparison tests', 'multiple sample comparison tests', and 'multisample comparison [tests]' are names which are sometimes used interchangeably.
'One way Anova' is an Anova test comparing the mean values for one variable between the groups/samples, to see if they are significantly different or not (it compares the groups 'in one way' or aspect).
'Two way Anova' compares mean values for two or more variables between the groups, and so on.
Sometimes an Anova test is made first to establish if there is a significant difference between results for the three or more groups, and this is then followed by a test such as Dunnett to provide additional information (eg: when an Anova test is used to tell you that samples do differ from one another, but doesn't tell you exactly which results are the ones which are different, and more information about this is wanted).
(as in)
http://www.texasoft.com/manual56.htm
"Multiple Comparisons Analysis
When you perform an analysis of variance test, the Ftest (main effects) usually is testing for the presence of a difference between factor means. However, if the test is significant, you do not necessarily know which of the several means can be considered significantly different. Some people have tended to perform multiple ttests to examine the differences between a number of means; however, this approach is incorrect since the pvalues associated with multiple tests are no longer appropriate. Multiple comparison tests are designed to allow you to perform differences between all possible pairs of means using an appropriate and controlled significance level.
WINKS allows you to perform four different kinds of multiple comparison tests. They are:
NewmanKeuls (Newman, 1939, Keuls, 1952)
Tukey (Tukey, 1953)
Scheffé (Scheffé, 1953; 1959)
Dunnett (Dunnett, 1955)"
Dunnett (Dunnett, 1955)
Dunnett's test is a specialized multiple comparison test that allows you to compare a single control group to all other groups. [others eg: Tukey (Tukey, 1953), Scheffé (Scheffé, 1953; 1959) ]
The test you use depends, in part, on your discipline. Some areas of research prefer one multiple comparison test over another. You should consult your literature to see which of these tests is most often used.
http://www.texasoft.com/manual56.htm
 Note added at 20 hrs 32 mins (20031109 14:00:53 GMT) 
typo.error:
 when I said: \"What is actually compared is one variable/characteristic (or more than one of the variables) in each group, compared with the corresponding variables in the other groups.\"
I meant: \" What is actually compared is the value for one variable/characteristic (or more than one of the variables) in each group, compared with the corresponding values in the other groups.\"
 Note added at 21 hrs 4 mins (20031109 14:32:58 GMT) 
Why do we need Anova (and Multisample comparison) ?
There are methods of comparing the mean values of two samples. Sometimes however biologists wish to compare the means of more than two samples. Suppose, for example, we have the length measurements from samples of three races of a species each of which lives on different islands A, B and C. It is possible to compare mean lengths [to see if there are significant differences] by ztests or,if the samples are small, by ttests. We would need to perform the test three times to compare AB, AC and BC. With the help of a calculator the task is not too daunting. Let us imagine instead that we wish to compare the means of seven samples. In this event no less than 21 ztests are required to compare all possible pairs of means. Even if the analyst has sufficient patience to work through this cumbersome treatment, there is an underlying statistical objection to doing so.
For example, if the \"P\"=0.05 (probability level is set at 5%) level of significance, a wrong conclusion will be drawn on average every 20 tests. In 21 ztests there is a good chance of at least one false conclusion. Anova overcomes these problems by allowing comparisons to be made between any number of sample mean values, all in a single test.
(With thanks to \"Practical Statistics For Field Biology\" Jim Fowler, Lou Cohen & Phil Jarvis  pub. Wiley 1998
Chapter 17 Analysis of Variance  Anova)
Reference: http://www.anselm.edu/homepage/jpitocch/anova/anovapage2.htm... Reference: http://www.texasoft.com/manual56.htm/
 DGK TI United Kingdom Local time: 12:48 PRO pts in pair: 401

  Grading comment Most thorough explanation, thank you ! 
 Login to enter a peer comment (or grade) 
Return to KudoZ list 
    X Sign in to your ProZ.com account...       KudoZ™ translation helpThe KudoZ network provides a framework for translators and others to assist each other with translations or explanations of terms and short phrases. See also: Search millions of term translations 