A sub-sample of the Labour Force Survey is the origin of this report. This survey is a quarterly sample of individuals living at private addresses in Great Britain. LFS started in 1979, however the current form of aggregated quarterly sample has been installed in 1984. The “60,000 respondents are identified through a random process of selection, stratified as to ensure representation of the main geographical regions of Britain” (LFS, 2016). Only half of the interviews were carried out over the phone, the rest face-to-face. The purpose of LFS is to provide information on the United Kingdom labour market that can afterwards be used to develop and evaluate labour market policies. (LFS, 2016)
My sub-sample consists of 6.000 cases drawn from the four quarterly samples aggregated together for the LFS 2005. Although my analysis will focus on three variables (ethnic groups, classes and weekly income) the sub-sample consists of eight variables: sex, age, marital status, weekly hours of work and country of birth. The sampling period of the survey is cross-sectional. The previous variables under my analysis had been recoded as it follows. The ethnicity group has been transformed in two categories: in “Black Caribbean & Black Africans” and in “Others”. Additionally, the classes have been transformed in three categories: “Service Class”, “Intermediate Class” and “Working Class”. The confidence intervals that I used in both of tests have the value of 95 %.
In the next section I am going to report the tests that I have conducted. Firstly, the t-test assesses whether there is an apparent difference between the means of my newly recoded ethnic groups and the weekly income variable and if this difference is likely to be reproduced in the whole population. The ethnic group is a nominal variable consisting of Black Caribbean, Africans and the others. The weekly pay is a scale variable consisting in the income expressed in pounds. The mean of black Caribbean, African was 395.20£ and the mean of the “others” was 483.55£.
Therefore, the average weekly income of those who are not in the black ethnic group is 88.35£ higher. The independent 1-tailed t-test is the appropriate procedure for finding out if the difference between the means of my two variables is indeed reproduced in the population as a whole.
My directional hypothesis: “that the average weekly pay of those who are other ethnic group is 88.35£ greater that those who are in the black ethnic group, in the population as a whole”. My null hypothesis: “that the difference between the two groups’ means in not reproduced in the whole population”. Levene’s test F-statistic was 9.629, with an associated significance level of 0.002, as this is less than 0.05 I assumed that group variances are not equal. My t figure was – 4.098, with 75.507 degrees of freedom and the 1-tailed significance was 0.000 after dividing the 2 tailed significance of 0.000 by 2. Because the associated probability,0.000 is less than the confidence threshold of 0.05 I rejected my null hypothesis and accept the hypothesis. Thus, I concentrated the analysis on the impact of the difference between my two groups’ means in the whole population. To demonstrate that my t-test is positive I have checked the four assumptions.
Firstly, the t-test examines the difference between two group means, my test compares the mean weekly income of black Caribbean, African and the means of those who are not, test met. Secondly, group samples are independent, as LFS survey is based on data that is aggregated over four quarters for each respondent and the ethnic group only permits a single category to be chosen there is no respondent that appears in both categories, assumption met. Thirdly, we have a directional hypothesis, test met. Lastly, we checked Levene’s test for equality and its assumption, test met. To conclude, I assumed that I have carried out a positive t-test.
As we can see, other ethnicities are paid with 88.35£ more weekly and with 353.4£ monthly than black people. Although results were collected in 2005, the wage gap is still a very contemporary problem that ethnic minorities have to deal with, not only in Great Britain, but in the whole world. Especially in the United States, African American women have substantial lower median annual earnings compared with Asian and American women. According to Major G. Coleman (2003), back in 1990-2000s the black men were earning weekly only 73.1% out of the white male income and the women were even more affected by inequality as they were only entitled to 62.5 % out of the weekly female income of 69.4%. In my opinion, education is one of the biggest factors for the wage gap around the world. A high majority of black people do not attend higher education institutions mainly because their families cannot support them financially as a lot of white family do. Although Clark K. and Drinkwate S. disagree that neighbourhoods have barely any impact on the future of young black people, I tend to say otherwise. (Clark K &Drinkwater S,2007) In the USA, most black ethnicities live in neighbourhoods controlled by African Americans, most of which are characterized by considerable drug use and violence. These people tend to develop little or no interest in getting highly educated or joining high profile careers. Consequently, two major factors for the wage gap between black ethnicities and others are the lack of education and work experience. (Berthoud, 2000)
To complete my analysis, I conducted the Chi-square test in order to test for an association between my newly recoded variables, the ethnic group and the socio-economical classes. Thus, the following contingency table was produced.
Ethnic Groups
Totals
1.Black Caribbean & Black African
2.Others
Classes
1. Service Class
0.5%
53.9%
54.4%
2. Intermediate Class
0.1%
10.5%
10.6%
3. Working Class
0.6%
34.4%
34.9%
Totals
1.2%
98.8%
100%
N = 6000
Table 1: Proportion of ethnicities in different types of socio-economical classes.
My hypothesis:” there is an association between a certain socio-economical class and ethnicity in the population as a whole” and my null hypothesis: “there is no association between a certain socio-economical class and ethnicity in the population as a whole”. I identified my chi-square statistic 6.523, with 2 degrees of freedom and associated probability of 0.38. As 0.38 is higher than the confidence threshold of 0.05 I accepted the null-hypothesis and rejected the initial hypothesis.
This limited my analysis, thus I restrained my observations about the distribution of socio-economical classes between the ethnic groups strictly to Great Britain.
For my results to be valid I showed that all four assumptions are met. Firstly, we are testing for an association between one or more categorical variables. Whether or not someone is in the black ethnic group is a nominal variable, and the class is an ordinal value therefore both variables are categorical. My chi-square tested for an association between these two variables, test met.
Secondly, the expected frequencies in each cell must be greater than or equal to 5. In my test the minimum expected frequency is 7.54, test met. Thirdly, each case occurs only once in the cells. Although respondents were re-interviewed for the LFS, results were aggregated for the four quarters of 2005. As there was no double-counting this assumption is met. Lastly, variables are independent, since neither ethnicity nor socio-economical class are calculated from another variable, this test is met. Therefore, I can state that I managed to attain a positive chi-square test with limitation to Britain.
The actual calculations of the number/percentages of people from the Black Caribbean, African communities, present in the three classes were essential. According to the row percentages, from all the people in the service class 0.91 % are black Caribbean, African. In the intermediate class 0.94% and in the working class we can find 1.71% of them. The column percentages, showing how many of the people in the black ethnic group are present in the previously mentioned classes are: in the service class 41.67 % of them, in the intermediate class we can only find 8.33% and in the working class 50.00%.
Considering these results, the presence of black ethnicities in the mentioned classes is alarmingly low in Britain. From all the employment positions in all the three classes the black minorities were only seen to be qualified for just 3.57% of them. Main causes for unequal distribution?
I believe that education intervenes here as well. In 2008 the high school dropout rate of Black Caribbean, African students was the highest in United Kingdom (Fleck,2008). This is the result of low family income and external influences. Moreover, in the rural areas of Britain, minorities have less access to wage employment than other ethnicities. However, Feldman and Steptoe state that in rural areas there are more opportunities for working class jobs which require less or unskilled workers, for which the minorities qualify. (Feldman, 2004)
Another cause is the restricted access of black people to health benefits. For instance, in 2005, researchers at the University of Essex reported that only one-fourth of the black women across Great Britain had a Pap smear in 3 years. Thus, women with lower social economic status are more prone to health risks which affect their careers. Therefore, unequal class distribution occurs (Brynin and Güveli, 2012)
In conclusion, the racial barrier is the main cause for unequal wage gap and class distribution among black Caribbean, African and others. Most privileges denied to black ethnicities are due to racial prejudices. University of Surrey reported that black minorities are 15% less likely to be employed than their British peers six months after graduation. Most employers tend to be biased in the selection process and discriminate black people by making interviews harder, for instance (House and Williams, 2000). Research has shown that black children are highly discriminated in the education system, either by their peers, or by teachers, who do not expect them to be as smart as white children (Fonagy, 2005). There are other branches where racial inequality occurs, but the length of my report restricts me to these aspects. However, the message is clear, as the British prime minister said: the “ingrained, institutional and insidious” racism is still present in the 21st century and we should all collaborate in order to fight it.