1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
  2. Hi Guest, welcome to the TES Community!

    Connect with like-minded education professionals and have your say on the issues that matter to you.

    Don't forget to look at the how to guide.

    Dismiss Notice

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

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

  1. alex_teccy

    alex_teccy Star commenter

    You should hit the reply button or use @ before a username to let the other poster know you’re responding or asking a question.

    I already outlined my relevant experience earlier in this thread.
     
  2. Stiltskin

    Stiltskin Star commenter

    Quick interpretation of the data gives me:
    124 out of 560 secondary headteachers are BAME in London in 2016 (about 22%, compared with ~50% of class teachers)
    513 out of 2120 primary headteachers are BAME in London in 2016 (about 24%, compared with about 40% of class teachers)

    London population was about 48% BAME in 2016
     
    alex_teccy likes this.
  3. Spoofer4114

    Spoofer4114 Lead commenter

    Okay so 22% and 24% are nowhere near 48% so when compared to the population that is a major under representation. The secondary statistic makes even grimmer reading given that the proportion of class teachers in schools more than matches the proportion of BAME in the general population (48%).

    All things being equal you would expect approximately 48% of headteachers to be BAME. For whatever reason things are not equal. Who can say why that is but as I stated earlier racism cannot be ruled out.
     
  4. Stiltskin

    Stiltskin Star commenter

    No you would not. It explains the reason in the report why you can not expect this.

    As they noted people are not employed directly into a head role. They have to start as a class teacher and work up. The pool of BAME in 2011 was smaller. Looking back to 2005 it was significantly smaller. How long do you think it take the average teacher to get to headship? It takes time for the effects to trickle through.

    The report shows that from 2011 to 2016 there was no significant discrepancy in promotion of class teachers to leadership roles. This means the proportion of those moving up roughly matched the proportion of the pool they came from. If this is the case then projecting forward from 2016 it should be around 30% in Primary and Secondary now.

    This is not as significant as the lack of women and BAME in Engineering in this country.
     
    alex_teccy and Rott Weiler like this.
  5. Spoofer4114

    Spoofer4114 Lead commenter

    Not significant at the 95% level. What about the 99% level. Would the differences have been significant at that level?
     
  6. Stiltskin

    Stiltskin Star commenter

    95% refers to the confidence level. 99% is a higher level. So where there were possible discrepancies there was a lower level of confidence that ethnicity was the contributing factor. I can not check at the moment, but I think they used an example of one group whose progression rates were slightly higher than white British, at around 60%. This is below the threshold so is not statistically significant and is likely that other factors influenced their progression. 95% is about as low as you would go.
     
    alex_teccy likes this.
  7. alex_teccy

    alex_teccy Star commenter

    Neither can it be ruled in.So why write it?
     
  8. Spoofer4114

    Spoofer4114 Lead commenter

    Not always. The 99% level is sometimes quoted and if the differences is significant at this level then race cannot be ruled out as a factor. In model 2 (comparing men and women) a treatment effect of 6.2% (progression to headship) was considered significant. The treatment effect for black teachers alone in model 1 is -6.5% (progression to leadership, page 51) which is a larger absolute difference so for this group this could be significant as well. There is no calculation for this particular confidence interval so I can't tell.
     
  9. Spoofer4114

    Spoofer4114 Lead commenter

    I have not said it can be ruled in.

    In statistical analysis there is no certainty. Doesn't stop people using stats as evidence when making a point. Guess you'll be asking Stilskin the same question because the stats he has shown don't rule race out as a contributing factor in teacher progression so why write it.
     
  10. Skeoch

    Skeoch Star commenter

    To try to explain the idea of the statistical test. In science, and even more so in social science, absolute proof is often unattainable. So we start by setting up a null hypothesis, in this case that there is no difference between the groups' progression. Then we see how often a random sample would produce an effect as big as the one we observe. If this is less than 1 in 20, the 95% level, we are reasonably sure that we can discard the null hypothesis. If it's 1 in 100, the 99% level, we are more sure still. The 95% level is the one normally used.
    As an analogy, if we are making decisions based on the toss of a coin, and I get two heads in a row, you won't suspect that my coin is biased. The chance of two heads is 25%. Now if I get 6 in a row, you will probably discard the null hypothesis that the coin is unbiased. The chance of getting six years is a bit under 2% so you'd probably be right and I might be cheating - but I might not, it might just be your bad luck.
     
    alex_teccy and Stiltskin like this.
  11. Spoofer4114

    Spoofer4114 Lead commenter

    Yes so in other words you can have a high level of confidence in your conclusion but you can't be certain. I have not said different.
     
  12. Stiltskin

    Stiltskin Star commenter

    They tried to find a discrepancy in the progression of staff based on their ethnicity but they couldn't. So in the data their may by some cases that are, but on the whole there isn't. In that data set. You suggest that racism is a reason that stops BAME becoming headteachers, but you have no proof.
     
    alex_teccy likes this.
  13. Stiltskin

    Stiltskin Star commenter

    If race is a factor that limits certain groups progression why do we not see it consistently across all teachers? If you are going to ignore statistics then you need to provide some other way to back up the suggestion that the lack of BAME head teachers is due to racism, because otherwise the suggestion that the actual reason is less BAME want to be head teachers is as equally valid
     
    alex_teccy likes this.
  14. Spoofer4114

    Spoofer4114 Lead commenter

    I have never said or suggested that racism is the reason that stops BAME becoming headteachers. I have always said that you can't rule racism out. Just saying that seems to be a problem for some as they seem determined to prove that you can rule racism out. There may be many reasons BAME don't become headteachers and racism could be one of them. If BAME teachers complaint that progression is a problem then it might pay to listen to their lived experience rather than try to contradict them with statistics that aren't proof of anything.

    It is ridiculous to me considering the history of this country in terms of race relations that BAME teachers claims of racism can be so easily dismissed. Having to have a law preventing people discriminating against minorities tells you all you need to know about some of the attitudes that exist in this country so it wouldn't surprise me if racism was at play. There are plenty of racist people in society and in all walks of life, including teaching.
     
    Missteach2010 and aalajamil like this.
  15. Skeoch

    Skeoch Star commenter

    @Spoofer4114 I think you've missed the point of the statistical test. It shows that there's no evidence of BAME progressing less well to senior positions. Once that's been accepted, then there's no need to discuss why the non-event didn't happen. I find this conclusion heartening.
     
    Stiltskin and alex_teccy like this.
  16. Spoofer4114

    Spoofer4114 Lead commenter

    No it doesn't the report points out that statistically at the 95% level there is no significant difference in career progression not that there is no evidence. There will be BAME teachers who will consider that they have evidence of their progression being prevented. At the 99% level this conclusion could be completely different anyway.
     
    aalajamil likes this.
  17. Stiltskin

    Stiltskin Star commenter

    99% is a greater correlation. If it doesn't meet 95% likelihood it can't meet 99%.

    "Statistical significance is the likelihood that a relationship between two or more variables is caused by something other than chance."

    In this case the rate of promotion by class teachers is linked to ethnicity.

    The highest correlation seems to be for Asians at about mid sixties, which showed their rate of progression may have been higher than other groups. But the correlation is so low then it could have been almost equally likely that other random factors caused this.
     
    alex_teccy likes this.
  18. Spoofer4114

    Spoofer4114 Lead commenter

    95% is a confidence interval not correlation. 99% confidence interval is smaller than 95% so there is more chance a result lies outside this. There isn't a scatter graph in sight. Where does it even talk about correlation in the report?
     
  19. Skeoch

    Skeoch Star commenter

    This data set is not of the type that can use a scatter graph. Nor is a correlation coefficient appropriate either. They are of use when you are looking at two continuous variables.
     
    Stiltskin and alex_teccy like this.
  20. Stiltskin

    Stiltskin Star commenter

    Correlation was wrong, as you and @Skeoch note. I may have been doing other things whilst trying to answer and tried to do it from memory.

    The Asian group showed the highest number of significant differences in the simulations run on their progression to a leadership role compared to white British, in 68% of the tests. To be confident this was not due to other factors not picked up in the census or in the randomness of sampling, it needed to appear in 95% of the tests to be be statistically significant that ethnicity was the reason for this. 68% is below this threshold so determined to be not statistically significant. You can see increasing it to 99% would not help as 68% is below 99%.

    The tests also showed that those that were identified as "mixed" appeared to do better at becoming headteachers compared with the white British group, but again this was not statistically significant.
     

Share This Page