Another view of that differential unemployment rate

If Andrew McCann has a key strength as a blogger, it’s that he loves numbers. He devoured yesterday’s story on high female Catholic unemployment rates with some gusto, and came out with a sound reasoning for the long term differentials.

  • Davros

    I’m not sure if I should be pleased or worried that his “math” agrees with the point I made yesterday đŸ˜‰

    “If Catholic women are 3.5 times more likely to be jobless in an overall ratio of 2 Catholics to 1 Protestant on the dole, a decent mathematician will conclude that the ratio of MALE unemployment between the two religions will be at virtual parity.”

  • carlosblancos

    1 If Catholic women are 3.5 times more likely to be jobless in an overall ratio of 2 Catholics to 1 Protestant on the dole, a decent mathematician will conclude that the ratio of MALE unemployment between the two religions will be at virtual parity.

    Please demonstrate how this is the case.

    2. The number of Catholic women entering the job market in Northern Ireland has been the fastest growing employment demographic since 1990. The Catholic baby boom of the early 1980’s is still feeding a disproportionate number of Catholics on the dole. Now that birth rates are practically the same as for non-Catholics, the figures of differential will disappear in years to come.

    Job growth has also been the fastest in the UK over the last 5 years. So therefore there should have been extra jobs for those extra Catholics who came along in their ‘babyboom’

    3. Nationalists are predominant in the rural west of Northern Ireland. Rural unemployment is a common factor across the whole of the United Kingdom. Rates of economic activity here are far greater in urban and metropolitan areas. Ulster is no different.

    True and probably explains some of it.

    What I find hilarious is that some unionists still claim that discrimination in employment does not exist. The figures are clear. No-one (except maybe Pravda) is suggesting that there is a large scale conspriacy, but discrimination is still the reality in some cases.

    The last comment on the post about moving south I find offensive and am surprised to see it on this site. It may not play the man, but it plays a group of men which is the same thing as far as I’m concerned.

  • CiarĂ¡n Irvine

    If Catholic women are 3.5 times more likely to be jobless in an overall ratio of 2 Catholics to 1 Protestant on the dole, a decent mathematician will conclude that the ratio of MALE unemployment between the two religions will be at virtual parity

    That only works if the workforce in both religions is 50% male and 50% female and the population as a whole is 50% Catholic and 50% Protestant…

  • Davros

    I don’t think it’s important when dealing with percentages CiarĂ¡n.

  • Occasional Commentator (was Commenter)

    CiarĂ¡n, (is there an &fada; or something for the fada?),
    It also assumes that males are as likely to be unemployed as females. This may be the shakiest assumption he has made.

    All he can say based on those two figures is:
    3upw + 2ucm = 4upm where u=unemployed, c/p is the religion, m/w = men/women

  • CiarĂ¡n Irvine

    Of course it is. Where did you learn your maths??

    Go on, sit down with a bit of paper and a pen and figure it out yourself with different population sizes and differing workforce participation rates by gender.

  • Davros

    Carlos – I’ll round up to 4 times for ease of mathematics

    if we have X as relative rate times

    if average of X ( m) and X ( f) = 2 times
    and X ( f) = 4 times,to achieve an average of 2 then X ( m) must be zero (Parity)as 4 + 0 divided by 2 = 2

    If X ( m) was 2.0 , then the average would be
    4 Plus 2 divided by 2 = 3 …

  • CiarĂ¡n Irvine

    OC: Most keyboards not set to US English will give the fada with CTRL+ALT+a. The HTML entity would be &aacute ;

  • CiarĂ¡n Irvine

    Dammit, HTML entity would be & aacute ; [without the spaces]

    Davros: parity ratio is 1, not zero.

  • Davros

    CiarĂ¡n – I learned maths same place as I learned manners. You need a refresher in both.

    In a situation where one group is above average then the other group must be BELOW that average…. that’s how averages work.

    Nobody is claiming that we are AT parity for Male Protestants and Roman Catholics.But we are very close. In 2004 the Equality Commission described the Catholic-Protestant balance in the Northern workforce as “very fair”.

  • Occasional Commentator (was Commenter)

    If males are much more likely to be unemployed than females then these stats show that catholic males are nowhere near parity. Sorry, Andrew is speaking out of his ass unless someone can tell us the male/female unemployment ratio.

    Ciáran,
    Thanks
    Use & if you want to show me an ampersand. Like so: á

  • Davros

    Sorry CiarĂ¡n – in this case zero …as in no difference.

    Fact remains – if one part of an equation is above average the other part must be below the average. That’s how averages work.

    If the average earnings of a man and a woman are 30 K and the man earns 40K the woman MUST be earning 20K …she Cannot be earning 30 K !

    So male RCs and protestants must be close to parity if the average of female RC is twice that of male and female RC combined.

  • Davros

    O C – try Alt Gr key with a – quite a few keyboards give you Ă¡,Ă©,Ă­,Ă³ and Ăº with Alt Gr.

  • Occasional Commentator (was Commenter)

    Davros, I’m working on a complete formula to show how weak Andrew’s point is, at least until we know the male/female ratio. But while I’m putting that together I want to point out that if males are much much more likely than females to be unemployed, then the females figures can pretty much be discounted, and the 2 to 1 ratio will be mapped directly onto the males.

  • fair_deal

    Carlos

    “What I find hilarious is that some unionists still claim that discrimination in employment does not exist. The figures are clear.”

    As Unionists are so often told when a there is a difference in favour of the Catholic community, disparity does not automatically mean discrimination.

    Also the latest research by the Equality Commission believes systematic discrimination does not exist in the Northern Ireland Labour market.

  • The Dog

    THis is all very interesting in a bland academic kind of way but if you look at the breakdown of some the stats used to create the noble index – on ill health for example or look at the economic inactivity ratesw or indeed some of the stuff produced by the Committee for the ADministration of justice on housing – or the patterns of investment to support inward investment or indeed the roads infrastructure, the railways then the BIG picture is far more interesting.

    There are clear geographical differences – west of the river Bann for example.

    Some of these throw of a pattern of policy of individual decisions that would support the more general argument about patterns of discrimination.

    Similarly the impact of the border is not just an anti-Catholic thing which was unionist and british policy for decades – it was just plain stupid.

  • Occasional Commentator (was Commenter)

    fair_deal, The Dog,
    You are quite right that it is difficult to interpret statistics correctly, but before we get into the subjective area of interpretation we must nail Andrew’s bad mathematics.

    We have to gather all the relevant numbers, sort out the maths, and then let rip with the interpretation.

  • foreign correspondent

    Maths is not my strong point so I’m not even going to try and join in the stats debate( the statistics component of Additional Mathematics GCE cured me of interest in maths for life…) But I’m feeling queasy for another reason-I just had a look at that Tangled Web page linked to in the post.
    Is that a site for the aspiring fuehrers who find the Daily Telegraph too leftwing or what? It has a link to Ann Coulter, for Nonexistent Deity’s sake ???!!!???
    Sorry, I think I need a lie-down after that.

  • George

    I’ll have a go using the figures from the end of 2004.

    unemployed = 37,000
    unemployed women = 11,000

    Let’s just take it as 60% Protestant and 40% Catholic workforce for both men and women which would mean a 60:40 split for the unemployed.

    2004 figures:
    Total number unemployed = 37,000 at a 60:40 P:RC ratio = 22,200:14,800

    26,000 men :11,000 women

    We have 11,000 unemployed women which at 60:40 P to RC would be 6,600P: 4,400C if the ratios were correct, which at 1P:0.666P.

    But the claim is 1P:3.5RC which is 1:.666×3.5 = 1:2.31. This is 7,676 unemployed Catholic women and 3,333 Protestant women.

    So there are 3,276 more unemployed RC women than expected.

    Let’s move on to total ratio which should be 22,200P:14,800C which is 1:0.666 but there is a 1:2 ratio here which implies 1:0.666×2 = 1:1.333.

    This gives us a new total figure of 15,858P:21,113C

    Now subtract the women’s figures:

    15,858-3,333P:21,113-7,676RC =

    12,525:13,437 = 1:1.07 ratio but as we should be 60:40 or 1:1.5 which means Catholic men are around half as likely again to be unemployed as Protestant men.

    I await my mathematical impalement.

  • Neal

    Fact remains – if one part of an equation is above average the other part must be below the average. That’s how averages work.

    Correct, but the question is how far men are below the average. If there are twice as many men as women in the pool of adults included in the sample, the figure for men will be closer to the average than the figure for women.

    If the average earnings of a man and a woman are 30 K and the man earns 40K the woman MUST be earning 20K …she Cannot be earning 30 K !

    Because in this case you have an equal number of men and women.

  • Occasional Commentator (was Commenter)

    I think it’s pretty clear that Andrew’s reasoning is dead, but can someone clarify what exactly the 3.5 figure is? Is it that there are 3.5 times as many RC women as P women on the dole? Or is it that the figure after taking a 60:40 P:RC ratio into account? And also the 2:1 ratio for the religion – I think that’s just a headcount, as opposed to “twice as likely”. Can this be confirmed?

    And is the 60:40 ratio between religions accurate?

  • Fraggle

    Well done George, nice to see that at least some people can count.

    Davros, night class algebra for you and Andrew I’m afraid.

  • Neal

    I’ll have a go using the figures from the end of 2004.

    unemployed = 37,000
    unemployed women = 11,000

    Let’s just take it as 60% Protestant and 40% Catholic workforce for both men and women which would mean a 60:40 split for the unemployed.

    2004 figures:
    Total number unemployed = 37,000 at a 60:40 P:RC ratio = 22,200:14,800

    26,000 men :11,000 women

    We have 11,000 unemployed women which at 60:40 P to RC would be 6,600P: 4,400C if the ratios were correct, which at 1P:0.666P.

    But the claim is 1P:3.5RC which is 1:.666×3.5 = 1:2.31. This is 7,676 unemployed Catholic women and 3,333 Protestant women.
    can someone clarify what exactly the 3.5 figure is?

    The 3.5 to 1 takes into account the 60:40 P:RC split, so Catholic women are 3.5 times more likely to be unemployed. If you were looking only at a headcount there are about 2.3 times as many Catholic women who are unemployed.

  • Neal

    Sorry about the repost of George’s calculations, which I was using to respond to OC…

  • Neal

    Whoops, I meant to post only –

    can someone clarify what exactly the 3.5 figure is?

    The 3.5 to 1 takes into account the 60:40 P:RC split, so Catholic women are 3.5 times more likely to be unemployed. If you were looking only at a headcount there are about 2.3 times as many Catholic women who are unemployed.

  • idunnomeself

    Well this is an interesting one. On the face of it Andrew and Davros (sorry mate) are wrong. It’s not that simple because you have to take into account the number of males and females in the economically active population.

    In other words stay at home mothers mean that many women who do not work are not unemployed, rather they are economically inactive.

    Whereas more men are economically active (IE either working or seeking work). Therefore it is quite conceivable that the figures given are correct.

    And there are lots of reasons for the figures, some of which have been mentioned above (I think the West/ East one is the most important, after all the adjacent areas of the ROI are the most economically deprived too).

    But the depressing thing is the way that some commentators wanted to find discrimination in figures which do not mean there is any. Such figures, because of the range of other factors in action, cannot be used alone to prove discrimintation (and Lord Laird should take note of that too).

    The Equality Commission doesn’t think there is any in NI any more. Is no Nationalist going to say ‘great boys, look what we wanted in 1969, we’ve got it!’?

    The same people who said of the Agreement that the problem was that the Unionists couldn’t tell if they won are showing that they have the same problem?

    And Davros, you’re right, clearly the issue to be addressed now is female unemployment. I am not surprised that this is being ignored, I learned long ago that an unhealthy obsession with Prods and Catholics makes so-called champions of equality blind to real inequality and real depravation..

  • Neal

    Whoops, I meant to post only –

    can someone clarify what exactly the 3.5 figure is?

    The 3.5 to 1 takes into account the 60:40 P:RC split, so Catholic women are 3.5 times more likely to be unemployed. If you were looking only at a headcount there are about 2.3 times as many Catholic women who are unemployed.

  • Davros

    If the average earnings of a man and a woman are 30 K and the man earns 40K the woman MUST be earning 20K …she Cannot be earning 30 K !

    Because in this case you have an equal number of men and women.

    Sorry mate – even if there are ten times as many men involved in calculating male average earnings of 40K or a tenth of the number of Men as women in the calculations – if men average 40K women cannot average the average đŸ™‚ Their average has to be less than the 30K mixed gender average.

  • Mark McGregor

    I’m too tired but the dedicated stat hunter could work out the unemployment differential in the 26 counties from this.

  • Occasional Commentator (was Commenter)

    Davros, you’re right there, but all that you are proving is that the unemployed male RC to
    P ratio must be less that 2:1. You cannot claim, as Andrew is, that’s it’s 1:1, or indeed 40:60.

    As others have shown you, it’s about 1.5:1

  • CiarĂ¡n Irvine

    Idunnno:

    But the depressing thing is the way that some commentators wanted to find discrimination in figures which do not mean there is any. Such figures, because of the range of other factors in action, cannot be used alone to prove discrimintation (and Lord Laird should take note of that too).

    Well, we can agree on something đŸ™‚ I was merely challenging the maths in Andrew’s post, not the implications.

    Personally, I think any overt discrimination as such, as it was understood in the 60s, isn’t really a significant factor any more. Sure, it might happen in isolated incidents here and there in particular firms but in general I think it’s a thing of the past. Any endemic continuing structural differentials in employment in the North are probably mostly down to a) historic underdevelopment of infrastructure in the west of NI b) the old problem of entrenched poverty: if no-one in your family for three generations has had a good job, yer not likely to see the value in education and work yourself and c) the impact of the border in reducing opportunity for spatial strategies for development in the likes of Derry-Letterkenny, Newry-Dundalk or even Armagh-Monaghan and Enniskillen-Sligo-Cavan.

    A coherent infrastructural investment strategy between north and south could help both jurisdictions, allowing economic development in the border regions leading to lower unemployment and an end to MOPEry like this. Saving both jurisdictions money in dole and increasing the tax take too.

    Course, the best way to co-ordinate that would be though Unification, but there are other ways that political Unionism is not interested in even though it might be the only way for them to save the Union long-term…

  • slug9987

    From Mark McGregor’s link I compute the following:

    Unempl InWork Sum Unempl as % of Sum
    RC 74680 833524 908204 8.2%
    COI 2175 29588 31763 6.8%

    Now (8.2-6.8)/6.8 = 20%

    Thus, in the 26 counties, the unemployment rate for Catholic men is 20% higher than the equivalent rate for Church of Ireland men.

  • slug9987

    “the old problem of entrenched poverty: if no-one in your family for three generations has had a good job, yer not likely to see the value in education and work yourself and “

    Peer effects have been shown to be important in numerous societies. Thus, if all your friends have jobs and are earning money, you feel you should too. Live is all about keeping up with your reference group.

  • Davros

    You cannot claim, as Andrew is, that’s it’s 1:1, or indeed 40:60.

    OC- I refer you to my original post yesterday đŸ™‚
    I haven’t claimed that it’s one to one.

    And I have offered a reasonable explanation for the differences – Female RCs face a double whammy –

    1)They face protestant employers discriminating against them because they are RC ( and there are more prod Employers , so there will be more RCs discriminated against than Prods – who in turn face discrimination from RC employers.)

    2) They face Male RC employers discriminating against them because of their gender.

  • aquifer

    I suspect some of the answer lies in the age profile of Catholic women, but historically poorer quality (less worthwhile to take) jobs and insecure (periods of unemployment) jobs for catholics will also be playing a part, as will welfare reforms that make it more worthwhile to work, (and therefore worthwhile to register as unemployed and actively seeking work)

    Entrenched poverty is a problem right across north and west belfast, with savings often non existent, giving little prospect of stepping on the property values wealth escalator.

    The good news is approaching full employment, which can make discrimination expensive for employers who need jobs doing. It would be good to keep that big wheel turning for the poor.

  • slug9987

    I don’t know if anyone has mentioned this point but could it be that catholic women, many of whom might previously not have wanted to work through child birth, have recently, because of the dip in the birth rate, started to supply themselves onto the labour market instead, creating a disproportionate number of the new arrivals in the queue of those seeking work?

  • Fraggle

    Mick, I object to your description of Andrew’s post as ‘sound reasoning’. I think George revealed it as the rubbish that it is. Andrew may love his numbers but his ability handling them shows that maybe he was better doing that arts degree.

  • Upsidedown

    Whilst it is heartening to see such lengthy debate on this topic, its more than disenchanting to see that no one has of yet mentioned what obstacles may perhaps exist to these women moving into work aside from that old familiar sectarian discrimination. This is definitely still a factor for some, though thankfully for less of us than it used to be, but for many more women in economically deprived areas a lack of skills that can be used in any kind of job other than the most menial and worst paying. What about the childcare aspect? The more kids you have the more you pay for their care and the more hours you’ve got to put in to cover it… All this mathmagician stuff is highly diverting but can we address the real issue???

  • Biffo

    Upsidedown,

    The real issue can be derived from from the following equation..

    6 into 32 won’t go

  • cladycowboy

    i got 5 and 1/3…

  • Occasional Commenter

    Davros, I think I did read your post yesterday and forgot it! I just got carried away because I was getting annoyed about Andrew’s logic and mistakenly lumped you in with him.

    Upsidedown, you say the mathematics is merely a distraction, but surely with an issue like this it’s important to get all the figures right first. I agree entirely that there are a lot of problems out there with discrimination et cetera, but we need complete and accurate numbers to inform the debate.

  • Davros

    Cheers OC – the very last thing I want is to be lumped in with ……

    People seem to be avoiding the sex discrimination component in this topic, Sectarianism is much easier to tackle. I was tempted to mention the relative levels of inherent sexism that exist in the main faiths, but thought that it would complicate matters even further!

  • Mick

    Fraggle, on mature reflection, plausible might have been a better term.

  • Neal

    Sorry mate – even if there are ten times as many men involved in calculating male average earnings of 40K or a tenth of the number of Men as women in the calculations – if men average 40K women cannot average the average đŸ™‚ Their average has to be less than the 30K mixed gender average.

    Of course it does, and I never suggested otherwise. The point is that without knowing the gender distribution of the sample it’s impossible to know where the number lies for men, except that it falls between 1 and 2. And there’s little basis for McCann’s claim that the figure for men is insubstantial.

  • Davros

    Neal – as ever there’s a degree of discussing at cross purposes.
    Cheers.

  • Joseyboy

    Andrew’s comments don’t tell us much we didn’t already know…about him.

    His mathematics is sub-‘o’ level ballderdash;

    He hates nationalists and catholics and has no wish to seek an accommodation with them; and

    He has zero sympathy for umemployed people of any religion or none.

  • Joseyboy

    Andrew’s comments don’t tell us much we didn’t already know…about him.

    His mathematics is sub-‘o’ level ballderdash;

    He hates nationalists and catholics and has no wish to seek an accommodation with them; and

    He has zero sympathy for unemployed people of any religion or none.

  • cxa493

    Ciaran are you from Derry?????