Smaller Schools = Bigger Schools

dambarude

A report out today from the Department of Education and Skills finds that small schools do not score significantly different to larger schools in measures of teaching and learning. See the report here .

Many would claim that smaller schools give more individual attention, and allow for better learning as a result. Others would claim the opposite. The measures in this report were fairly rudimentary, but it would appear given the amount of schools involved that there is little difference between the two types of school in terms of assessment, parent ratings etc.

Victor

The innumerate gobsh!tes.

Data from Parent Questionnaires
Schools with 49 or less pupils
Teaching is good in the school.
Mean 1.37231738218109
Std. Deviation 265475601172637 (no decimal point)

It suggests some schools are perceived as billions of times better than others.

dambarude

I would imagine that was meant to read .265475601172637 given the mean score involved. A typo rather than a statistical error I would imagine:rolleyes: It gives the correct figure on page 34.

byhookorbycrook

In fairness,parent questionnaires at WSE time are hardly objective data.Likewise , inspectors are not likely to be uniform in their "judgements."

dambarude

True byhookorbycrook, but do you think that might be balanced out by the fact that data is taken from several hundred schools, and inspector biases in a small school would be equally apparent in a larger school?

Victor

Who honestly gives any number to 14 decimal places, especially in statistics?

dambarude wrote: »

True byhookorbycrook, but do you think that might be balanced out by the fact that data is taken from several hundred schools, and inspector biases in a small school would be equally apparent in a larger school?

No, it's opinion. Subjective, not objective. Wouldn't it be more useful to see how well students are actually doing, e.g. give them an exam and adjust for intake.

Given that

• The number of inspectors is limited, there is a risk of group think.
• Notwithstanding this, each inspector is likely to have a somewhat limited variety in the schools that they inspect and individual inspectors may be unable to compare small rural schools with large urban ones.
• Given that schools typically use geographic catchments, it is difficult for parents and pupils to compare schools.
• School size is correlated with social grouping
• In smaller schools in particular, school quality may be directly linked with the performance of 1-2 individuals directly known to the survey respondent. The pupils may be dependent on one individual for up to 8 years of hteir education.
• Only a limited number of subjects are considered - reading, maths, planning and assessment.
• There is a lack of definitions - what is ERC and how is "assessment" defined?
• A modest number of small schools appear to have been surveyed (there are about 3,000 primary schools).

the exercise seems unduly limited in scope and analysis.

Were the schools picked at random?

No comparison is made between schools with (a) very small numbers (0-30 pupils - less than one nominal class size) (b) merged classes, i.e. more years taught than teachers (c) medium sized schools with one teacher per year (d) larger schools with more teachers than years taught (200+ pupils)

dambarude

All good points. Re: the decimal places, all of the tables seem to have been copied directly from SPSS. It's clear from the formatting. It's very lazy, and would appear to me to have been done in a rush.

It's all subjective data, but I would think that is all they have to go on. If they had access to, and started comparing standardised test scores they would be opening a very large can of worms.

The sampling wasn't done for this analysis in particular, but rather has been drawn from incidental and WSE inspections. So the data wasn't collected with this report in mind. I'm not sure how it's decided who is inspected, and when, but it appears to happen to every school every few years. So I'd doubt this was targetted. Given there was a sample size of nearly 1000 out of just over 3000 primary schools, I would have thought that sampling wasn't much of an issue.

By the way, I'm in no way endorsing the results. I haven't had time to read them in full, only the headline figures. You'd have to read the incidental inspection reports and WSE reports separately to get a good idea of what the reported figures actually mean, and how they were collected.

dambarude

Reply to your edit.

Victor wrote: »

No comparison is made between schools with (a) very small numbers (0-30 pupils - less than one nominal class size) (b) merged classes, i.e. more years taught than teachers (c) medium sized schools with one teacher per year (d) larger schools with more teachers than years taught (200+ pupils)

These would have been the most useful stats to have gotten out of the report. There's a big difference between schools with 49 pupils and with 10 or 12. The class distribution and number in each would be very different. They might not have had enough cases of schools of this size to allow for valid statistical comparison with other categories, but that's just a guess.

byhookorbycrook

Just to comment on inspector judgement,they can't even agree standard formats for dips. Also, no two schools are alike,so it's not just small vs big, there are rural DEIS too, some areas have higher numbers of children needing extra support and so on, so the whole process is entirely flawed.

dambarude

byhookorbycrook wrote: »

Just to comment on inspector judgement,they can't even agree standard formats for dips. Also, no two schools are alike,so it's not just small vs big, there are rural DEIS too, some areas have higher numbers of children needing extra support and so on, so the whole process is entirely flawed.

Thanks byhookorbycrook. Would you think that there is a difference between 'big' and 'small' schools then?

byhookorbycrook

ERC is the educational research commission, I'd imagine?As to incidentals, a young teacher in a local school was severely taken to task for "the wrong kind of clock" in her room on an incidental, so might have been deemed unsatisfactory by that gobdaw of a cigire, it's all too subjective to be of any use.Same cigire told the school that their standardised tests results should have been better for the "kind of school they were?!??", no objectivity there!

dambarude

I think the ERC is the Educational Research Centre in St. Pats.

byhookorbycrook

There are too many variables in the mix to be able to say small vs big.

Take a small school in an area where the parents are mainly middle class with an interest in education vs a small school where there is not the same interest. Take a bigger school that has seen very few children with additional needs join the ranks vs a school where up to 40% of the school population may not have English as a first language.
Take the survey going to a parent with an axe to grind due to some perceived slight as opposed to going to another parent who thinks teacher=deity, it's all too vague and there are too many combinations.

I'm sure the DES would love to be able to "prove" that small schools are a disaster.Lies,damned lies and statistics and all that.

byhookorbycrook

dambarude wrote: »

I think the ERC is the Educational Research Centre in St. Pats.

Ooops, centre, not commission, well done, Damba. For the first time ever we had to sign a form to say we were qualified to give the Drumcondra tests this year. The usual q "what do you want these tests for " bit always makes me snigger, like someone is going to write "I'm really a parent and and pretending to be a teacher so I can coach my child in this test so I can then brag to everyone that little Chardonnay is a STEN 10."

dambarude

I've seen parents first hand test their children on standardised tests (the one not used in the child's particular school).

Do you not need more proof than that to obtain the tests?

byhookorbycrook

Roll number and signature of a principal,but I know one eejit who decided to give out the tests to the parents to take home :rolleyes:
Happy to say that the inspector's "subjective view" in this case meant the same eejit is no longer in any school

dambarude

Wow... Sadly I think they are going to be misused in even more cases in future given the emphasis being put on standardised tests in the Num and Lit directive. It will surely lead to more teachers teaching to the test.

byhookorbycrook

But if we all teach to tests, results go "up" and Ho Chi Quinn will be able to announce to the media how wonderfully well his strategy has worked....

dambarude

I hadn't thought of that. It's a scary prospect. I think we'll need some PISA results to confirm it though - those are the results which they seem to place the most store in

Victor

byhookorbycrook wrote: »

There are too many variables in the mix to be able to say small vs big.

Actually you can. However, the way the report does it is a bit cack handed, not only because the distinction used is rather arbitrary, but that it doesn't control for other factors like geographic location.

However, big may suit some locations and small others. I imagine an isolated small school could suffer excessively in the event of persistent absence / failure by a teacher, whereas in a 16-class school, there is likely to be sufficient number of spare teachers / teaching assistants to maintain continuity. A small school closer to home is useful in other ways. Amalgamating schools may cost money if new buildings have to be built, but would save money if spare capacity existed in one of the schools. A small school will likely struggle to deal with multiple language backgrounds, whereas a large school could see a disproportionate number of students affected by poor management.

ShamoBuc

Is this thread for mods only:pac:

I wouldn't pay too much attention to it. Alot of reports are compiled but not all are released with a bit of 'fanfare'. Depending on the viewpoint of the RQ and co certain ones seem to get extra attention.
The one on Patronage has come out aswell and is a fascinatong read.
Anything that helps iron out any problems they might have in 'alterating' schools seems to be very popular at the moment:rolleyes:
And at the same time as all the unions are having their agm's aswell.

echo beach

I don't know much about statistics but I have to use them occasionally and I do know that you can't just throw figures into a software package and come out with something meaningful.
As has already been pointed out the number of variables, both those included and those not included, is too great for the small sample size.

How 'good' a school is will always be subjective. Parents want different things. Some value academic sucess, others sporting sucess, others look for a more 'holistic' approach to a child's development. Most want schools to do far more than is possible given that home and wider society have such an important role in 'education' in the broadest sense of the word.

There are 'good' small schools and there are 'bad' small schools, just like there are 'good' big schools and there are 'bad' big schools. What makes a school 'good' or 'bad' isn't its size or the physical enviroment, it is the quality of the teaching and the leadership of the principal and management.

nice_guy80

very easy for a teacher to spend two weeks before the test practicing similar type questions in order to boost the results of pupils

it is what happened in NI for years. prepare for the test

would parents be happy then ith better result?
or the BOM?
or Ruairi Quinn?