# BAME SLT members of staff could help alleviate issues in some schools.

Discussion in 'Education news' started by mark_oviri1, Mar 15, 2020.

So t-test were performed on samples

Picking 100 random samples of teachers from the control group that matched the teachers in the treatment group gave different results – sometimes the t-test comparing means for each sample showed that the treatment group was statistically significantly different from the control group and sometimes not.

It does not state at what level of confidence the significant difference was calculated for these t-tests. Different levels can produce different conclusions. What is significant at one level is not significant at another. The 95% refers to the number of tests needed to conclude statistical significance (seems a particularly high bar) not the confidence interval for each particular t-test.

You will note that all this does anyway is show statistical significance. It is not proof of anything.

2. ### StiltskinStar commenter

So write a paper to explain how they are wrong then!

3. ### SkeochStar commenter

No - you've misunderstood the methodology of the research and of t-testing,
"Not significant" unless qualified means not significant at the 95% level. 95% does not refer to the number of tests needed. 95% isn't a high bar - it's the normal level that we work at.

What is this all about then on page 73?

For us to be confident that the outcome from the treatment group was statistically significantly different from the control group we required 95% of the t-tests for these samples to be significant.

I didn't say they are wrong, they do show no statistical significance at the level they have used, I am saying they do not prove that race is not significant in career progression. To be able to do that you would have to be able to read the minds of people making the promotions.

6. ### drvsStar commenter

I see you've recognised that localised grouping has statistical significance.

Stiltskin and alex_teccy like this.

I never said there wasn't. You can provide the quote where I said this?

Last edited: Apr 27, 2020
8. ### drvsStar commenter

thin_ice likes this.
9. ### StiltskinStar commenter

Yes it shows no statistical significance in disparity of the promotion of teachers by ethnicity. Race may be significant in some career progressions, but they are looking at the mean rate for groups not individual cases. When looking at groups of teachers who all other things being equal, then on average there is no difference.

So if, on average, there is no difference in progression of teachers based on ethnicity it suggests that it is not this that is the reason for the disparity of the number BAME in leadership positions. Which is what you initially questioned. If there is less BAME teachers though, then there would be a small proportion of them being promoted. This would be evident in regional disparity - there are more BAME in leadership positions in London because there is a bigger pool of BAME teachers to start with.

Wrong. I have never questioned the conclusion of the report at the confidence level given. I said if different confidence levels were used you could get a different conclusion in terms of statistical significance. I have always said that racism cannot be ruled out when talking about career progression of BAME. It cannot be ruled out and statistics cannot prove that it can.

Do you think the report proves that race plays no part in the career progression of BAME teachers? And when I say prove I mean it can be stated with 100% certainty that race is not a factor in the career progression of ethnic minority teachers.

Last edited: Apr 27, 2020
11. ### SkeochStar commenter

No. It doesn't prove things either way. What it does is to fail to disprove the hypothesis that there is no difference between the groups.
And as it's working at the 95% level, any lower level (90%, say) wouldn't be an acceptable approach as the result might come about by chance. Clearly if it's not significant at 95% it cannot be significant at 99%.

Stiltskin likes this.
12. ### thin_iceOccasional commenter

drvs likes this.
13. ### StiltskinStar commenter

There is a difference between probability and possibility. What you are asking is to prove the non existence of something, which in this case is not possible. There are too many variables to account for.

What the report is telling me is it in not very probable (EDIT - based on the data they have) that the difference in the number of headteachers who are BAME or white British is due to the lack of progression of teachers because of their ethnicity. That the reason is probably due to some other factor(s). It is still possible that some teachers find they don't progress because of their ethnicity (and this can be true even if they are white British - you can not rule that out either). The report does show that a teacher's gender will probably effect their chances of progression to leadership. Similarly it is possible that some teachers this won't be a factor.

So let's flip this around and you show the proof that race plays a significant part in the career progression of BAME teachers?

drvs, alex_teccy and Rott Weiler like this.

Why do I need to do this when I have never claimed this was the case?

You still haven't answered this question either.

Do you think the report proves that race plays no part in the career progression of BAME teachers? A simple yes or no will suffice.

At last. Took a while but we got there.

Well done.

Last edited: Apr 27, 2020

17. ### drvsStar commenter

You failed to be able to see how localised grouping could account for any disparity, then later you used localised population figures to argue a disparity.

You then failed to see the link between the two quotes I had already posted and quote challenged me, now you're going for a semantic challenge and patronising me.

If you've set out in this thread to prove that you're not as bright as you think you are, I would say that you are making a great success of it. Keep up the good work.

alex_teccy likes this.

This is what you first said

I have never said anything about not recognising statistical significance in localised groups. I was saying I could not see how disparity in localised groups could affect the overall statistics given in the Guardian article.

We have since moved onto another report.

Do keep up.

Easy with the insults as well. Anyone would think you were getting flustered.

19. ### StiltskinStar commenter

If you don't think that then why bother to ask...
I did answer, see previous post. Your wording of the question suggests you possibly misunderstand what the report is showing.

Tbh I'm not sure you know what point you are trying to make either.

alex_teccy likes this.