This cross-tabulation table above shows that generally there is a positive correlation between those who complete full time education and disagreement with the statement that” asylum seekers should be sent home”. There is one clear anomaly however. This is exemplified by the participants who left school at 17 yet ‘strongly agree’ and ‘agree’ to an equal or larger extent with the statement in comparison to those who left school at 15 and 16. However it could be seen that it is not to do with age of completing education but rather the age the participants are when undertaking the survey. It would be interesting to see a comparison between current age of participants and the age they finished education to see whether it was increased education that changes viewpoints or just the society participants grew up in. These cross tabulations do not explain the causality between the two variables. However, the use of a source as the ‘British Electoral Study 2005’ as well as the great significance score of p<0.01 act to give the results credibility as reliable and definite.
The correlation table above addresses the potential relationship between negative attitudes towards asylum seekers and negative attitudes to immigrant in general. The two variables that are examined are the views on whether ‘asylum seekers should be sent home immediately’ and whether ‘immigrants increase crime rates’. The positive correlation of 0.551, which should be seen as reasonably strong suggests that there is a moderate positive link between the variables. The fact that the correlation coefficient is followed by two asterisks, shows the result is very significant. This is represented by the chance of randomness (p), being lower than 0.01, meaning that there is less than a 1% chance that the results are random. The ‘n’ shows us that 703 people participated, however we are not told the possible amount of people who would be eligible for the survey which would be represented by the character ‘N’.
The regression table above acts as a tool to decipher the effects that the independent variables of ‘age’, ‘income’, ‘age of leaving school’, and ‘gender’ have on the dependent variable which is agreement with the quote ‘asylum seekers should be sent home immediately’. The independent variable which has the largest effect on the dependent variable will be the one with the greatest standardised coefficient, and to make it valid, the lowest significance score. Statistically the ‘age of leaving school’ has the highest standardised coefficient at 0.268, as well as the lowest significance score of 0. This has clear similarities with the correlation that is present in the cross tabulation between age of leaving school and agreement with the statement ‘asylum seekers should be sent home immediately’. This shows that ‘age of leaving school’ has an important effect on the dependent variable, and is a result which is clearly not random. The alternative independent variables have an insignificant impact on the dependent variable. Moreover the fact that these independent variables all have significance scores greater than 0.4 renders them largely useless to study. Another weakness in this data is shown by the fact that the R2 is just 0.073 which means that the model explains only 7.3% of the dependent variable. This is a low percentage and one that could be improved with the addition of more independent variables.
To conclude, when analysing what we have found out about the dependent variable it appears that very little knowledge has actually been gained. To summarize, the data is not precise enough and although this may be seen as a general characteristic of quantitative analysis it is exacerbated in these statistics due to a lack of specificity. The fact that 20% of people opted for ‘neither agree nor disagree’ as their answer suggests that the question was not clear enough. For example, it is not explained what actions ‘strongly agree’ or ‘agree’ entail and so for one person the actions that might be categorised by the answer ‘strongly agree’ might be categorised by just ‘agree’ for another participant. Moreover there is no room for participants to distinguish between a deserving immigrant who is possibly fleeing persecution and an undeserving immigrant who merely seeks government welfare which could also explain why 20% opted to ‘neither agree nor disagree’. Furthermore as has been addressed earlier the data fails to provide any causality, therefore although we learn that school leaving age has an impact on the attitudes towards asylum seekers, it is left open to why this is the case. Not only does the data not provide causality, but the fact that the data is collated only from 2005 makes it susceptible to specific societal conditions that may have been prevalent at the time which influenced people attitudes
Bibliography
Blakie, N. Analyzing Quantitative Data: From Description to Explanation (London: Sage Publications, 2003)
Connolly, P. Quantitative Data Analysis in Education: A Critical Introduction Using SPSS (London: Routledge, 2007)
Harrison, L. ‘Part I Quantitative Research’, Political Research: An Introduction (London: Routledge, 2001)