Research Report

Introduction

This research report will look at results and data from the Northern Ireland Life and Times Survey (also referred to as NILT)

Based on the results from the regression model, my research question focuses on whether religion plays a role on how much a person earns on their annual personal income. I will also look at whether sexual orientation impacts personal income. My hypothesis will be that Catholics are more likely to earn more than Protestants in Northern Ireland, and that religious category does in fact play a role in how much someone earns, as well as sexual orientation.

Data and Method

This data set was collected by two universities in Northern Ireland who worked together to collect data on the people of Northern Ireland. The research centre set up by both of the universities is known as ARK, it is them who conducted the research. This survey was published in 2012, although ARK claims to run many surveys measuring attitudes in Northern Ireland on an annual basis. The interviews for this survey took place over the span of three months, from the 1st October 2012 till the 10th January 2013.

The NILT survey began in 1998 by the ARK research centre in order to measure the opinions and overall behaviour of the population. The survey also provides a good opportunity to measure how attitudes have changed over time in a real way backed up by evidence.

The data found in the 2012 Northern Ireland Life and Times survey shows the general attitudes of people in Northern Ireland on a number of topics, as well as the general background variables such as age, sex, area, religion, etc.

The 2012 survey that I will look at contained 5 complete modules in the dataset I looked at. Those were: Background information on the respondents, Community relations, Attitudes to ethnic minorities and migrant workers, Political attitudes, and LGBT issues. In the report it notes how a sixth module was used in the survey but was not included in the document I read, this was on Autism.

The survey is split into three types of variables:

Section 1—Background/common variables

This section details the questions asked that are to do with background. Some of the main questions involve variables in relation to income, age, sex, family, work, income, religious affiliation, personal relationships, social class, education, and even whether the respondent has access to the internet. This section deals primarily with nominal data.

Section 2: Good Relations variables

This section looks at measuring the attitudes of respondents in relation to their relationships with other groups. This includes: opinions on issues regarding racial prejudice and ethnicity in general, religious prejudice with somewhat of a focus on the Catholic/Protestant divide in the region, migration, Unionism and Nationalism, and more general things in relation to sectarianism. This section includes ordinal data in the form of rankings as answer options. It also involves nominal and scale.

Section 3: LGBT Variables

The third section focuses on questions in relation to LGBT issues. The measure of data in this section is primarily ordinal, however, it includes scale and nominal aside from that. Throughout the three sections, the questions remain with those three measures consistently.

The questions in this section relate to attitudes towards people in society that identify as being a part of the LGBT community. This includes: how comfortable respondents are towards gay men/lesbians, how much LGBT should be allowed to express themselves, if Same-sex couples should be able to get married, adoption issues, and Trans-related issues.

Data collection & size

The sample size in this size was 2126 eligible respondents. The researchers used the method of random sampling in order to find respondents for the survey. They selected a total of 2350 addresses from the Postcode Address File database of addresses, with 224 of the selected addresses being illegible for the survey.

The next stage in the selection process involved working out how many people lived in each household at each of the randomly selected addresses, and choosing one individual to take part in the survey. The report notes that it used a method which they referred to as the ‘next birthday’ rule. This involves asking each of the members of the household over the age of eighteen who was next to have a birthday.

There are certain advantages and disadvantages with using the method of random sampling when conducting a survey for research.

The main advantages of this method are:

Each member of the larger population has the same chance of being chosen therefore it is equal

It is the most frequently used method of sampling and therefore it can be argued that it is the most reliable.

Researches are able to get general view of the population’s attitudes [Moss PhD, Rosenzweig MS and Litman PhD, n.d.]

The main disadvantages are:

It could be costly when trying to contact a large group of people for surveying [Moss PhD, Rosenzweig MS and Litman PhD, n.d.]

The process of collecting data from a large group selected randomly will take a long time compared with other methods [Moss PhD, Rosenzweig MS and Litman PhD, n.d.]

Response

Of the 2126 eligible households randomly selected to be part of the survey, only 1204 face-to-face interviews took place when it came down to data collection. This means that the survey had a 57% response rate as noted in the report. 32% of randomly selected respondents refused to co-operate and 12% made no contact with researchers. With just under half of chosen respondents not taking part in the survey it can mean the results of the survey are less accurate as a whole. There is a risk that the attitudes recorded do not reflect the larger population of Northern Ireland.

Part 2: Results

The regression results are shown below:

This summary shows the average personal income for someone in Northern Ireland according to the survey is £11,960 and that the highest income recorded was £75,000.

Bar Graph

Religion:

The bar graph below gives a visual account of the findings:

1 is referring to Catholics

2 is referring to Protestants

3 is referring to No Religion

As you can see, Catholics seem to earn more at every category of personal income. Interestingly, Catholics and Protestants are roughly the same when it comes to earnings over £45,000. The lower the amount of income the more likely they are to be Protestant. This could be because the survey interviewed more Catholics than Protestants and due to the potential failings of the research method undertaken it may not reflect the whole of society in Northern Ireland. In order to produce a bar graph I used the ‘ggplot’ command which is used in making graphs

Sexual Orientation:

The following bar graph shows the correlation between personal income and sexual orientation:

1 is referring to Heterosexuals

2 is referring to Homosexuals

3 is referring to Bisexuals

4 is referring to Other

The main conclusion to be taken from the data is that the number of heterosexual people surveyed vastly outnumbered those that were homosexual, it can be said that this does reflect the number of Homosexual/Bisexual people in society. According to statistics from the ONS, in 2019, 93.7% of people in the UK identified as Heterosexual [Office for National Statistics, 2021]. The majority of people who do identify as Homosexual earn the average personal income, this is evidence that sexual orientation may not play a role in personal income in Northern Ireland. However, perhaps a survey on a larger scale would provide us with a different conclusion as there would statistically be more LGBT people surveyed.

Limitations

There were a number of limitations I found while conducting my research and preparing my report. The main one was my inability to grasp the Rstudio software used to prepare the report. I encountered many issues with errors and was unable to fix them. Another limitation is the number of people surveyed versus the overall population of Northern Ireland which does hinder the reliability of the results of the survey.

Conclusion

To conclude, the following research from the 2012 Northern Ireland Life and Times survey, conducted by the ARK research centre shows that Catholics do, in fact, earn more than Protestants in terms of personal income. The lower down on the income scale, the more likely it is that the person is Protestant rather than Catholic. Therefore, proving my original hypothesis to some extent. However, there is no reason to suggest that religion plays a role in a persons income. There is also no real evidence from the survey to suggest that sexual orientation plays any kind of role in personal income as the data shows that those that identified as other than Heterosexual earned the median amount of personal income.

References

Moss PhD, A., Rosenzweig MS, C. and Litman PhD, L., n.d. Pros & Cons of Different Sampling Methods | CloudResearch. [online] CloudResearch. Available at: https://www.cloudresearch.com/resources/guides/sampling/pros-cons-of-different-sampling-methods/.

Fogarty, B., 2019. Quantitative Social Science Data with R. London: SAGE Publications Ltd.

Office for National Statistics, 2021. Sexual orientation, UK: 2019. p.2.

ARK. Northern Ireland Life and Times Survey, 2012 [computer file]. ARK www.ark.ac.uk/nilt [distributor], June 2013.