Hi, Good Morning.  I saw your question and its very interesting.

Statistical control, used when experimental control is difficult, if not impossible, can be achieved by measuring one or more variables in addition to the independent variables of primary interest and by controlling the variation attributed to these variables through statistical analysis rather than through research design. The analysis procedure employed in this statistical control is Analysis of Covariance (ANCOVA).

1.  ANOVA that provides a way of statistically controlling the (linear) effect of variables one does not want to examine in a study. These extraneous variables are called covariates, or control variables.

2.  Covariates should be measured on an interval or ratio scale. It allows you to remove covariates from the list of possible explanations of variance in the dependent variable.

3.  It does this by using statistical techniques (such as regression to partial out the effects of covariates) rather than direct experimental methods to control extraneous variables.

4. ANCOVA is used in experimental studies when researchers want to remove the effects of some antecedent variable.

For example: Pre-test scores are used as covariates in pre-test & post-test experimental designs.

5.  ANCOVA is also used in non-experimental research, such as surveys or nonrandom samples, or in quasi-experiments when subjects cannot be assigned randomly to control and experimental groups. 

6.  Although fairly common, the use of ANCOVA for non-experimental research is controversial.  For further, kindly refer the paper wrtten by Vogt et al. in 1999.

Thanks.

.

ANCOVA is generally used where the main interest are categorical predictor variables, and you can control the effect of interfering variables - either categorical or continuous.

1. My question is: should these differences be statically significant or not?

It depends on the measurements you have. Sometimes, the baseline measurements are different between two groups whereas sometimes they aren't. It just depends on data you have collected.

2.When there is a difference between pre-test scores, but it isn't statically significant, can I consider pre-test scores as covariate yet?

You should decide how important (scientifically) the pre-test scores are. Statistical significance isn't the be all or end all (A p-value shouldn't be taken as a gospel and final truth). It is good to adjust them whenever possible. If they are not significant (or not scientifically important), then they can be ignored. However, you can always examine how important your baseline measurements are. You can add the pre-test scores in your ANOVA (or regression model) and assess how the Sums of Squares (or maximum likelihood estimates) for other variables changes and decide on whether to keep them or leave them out of the model.

HTH.

Hi, Good Morning.  I saw your question and its very interesting.

Statistical control, used when experimental control is difficult, if not impossible, can be achieved by measuring one or more variables in addition to the independent variables of primary interest and by controlling the variation attributed to these variables through statistical analysis rather than through research design. The analysis procedure employed in this statistical control is Analysis of Covariance (ANCOVA).

1.  ANOVA that provides a way of statistically controlling the (linear) effect of variables one does not want to examine in a study. These extraneous variables are called covariates, or control variables.

2.  Covariates should be measured on an interval or ratio scale. It allows you to remove covariates from the list of possible explanations of variance in the dependent variable.

3.  It does this by using statistical techniques (such as regression to partial out the effects of covariates) rather than direct experimental methods to control extraneous variables.

4. ANCOVA is used in experimental studies when researchers want to remove the effects of some antecedent variable.

For example: Pre-test scores are used as covariates in pre-test & post-test experimental designs.

5.  ANCOVA is also used in non-experimental research, such as surveys or nonrandom samples, or in quasi-experiments when subjects cannot be assigned randomly to control and experimental groups. 

6.  Although fairly common, the use of ANCOVA for non-experimental research is controversial.  For further, kindly refer the paper wrtten by Vogt et al. in 1999.

Thanks.

.

If your initial aim was to correct for a possible baseline difference, you should stick to that decision and leave the baseline value as a covariate in your model. Irrespective its significance.

 I have one more question.

My consulting adviser said we can't use covariance method when there are more than 2 time points. But I'm not sure about that again!

What's your idea about that?

hello, I need article use Ancova please

You may possibly find Andy Field 3rd Edition Ch 11 (ANCOVA) page 398 where he talks about anxiety/depression as a confounder helpful.