On the other hand, official statistics can be useful in providing relevant statistics in areas such as household income. They cover large populations and are therefore representative, which is useful when conducting a large, quantitative study. This information can be linked to educational achievement by studying income alongside pass rates in schools in poor/wealthy areas.
However, marketisaition increased the importance of money in determining success. Smith and Noble (1995) suggest it produced an increased polarisation between poor/underfunded, underachieving schools in low income areas and successful, well resourced schools in affluent areas. When studying this, sociologists must combine official statistics with more qualitative data; so, although (from a positivist perspective) this offers a complete structural picture (as location and quality of schools can be discerned through OFSTED reports, which draw on the advantages of being well funded, objective and reliable), the information provided by official statistics must be correctly analysed and collated in order to form a credible argument.
They benefit from the defining strength of most secondary sources: official statistics are inexpensive and readily available and accessible. Census data is the product of a £400 million project, and sociologists would otherwise be incapable of accessing such data. The inclusion and consideration of such large scale studies adds balance to a study, with standardised, nation-wide research lending the text credibility and a more objective, all encompassing structuralist element.
Nevertheless, a sociologist must draw together numerous statistics and sources when developing an argument. Official statistics can be tinged/distorted with bias, depending on who has funded/conducted the research. For example, investigations into river pollution in Canada have been criticised for favouring the backers of the independent firm undertaking the investigation – in this case, the oil company funding the research. Such distortion can often invalidate entire studies, with results lacking the neutrality and objectivity to be considered truly accurate and reliable.
However, they lack the depth of unstructured interviews; interpretivists argue that such an in-depth approach is required to understand the complex issues involved with education, social class and poverty → linked to 'soft' statistics' which are unclear and interpretivists claim that these are social constructs reflecting the ideologies of those conducting the study, rather than valid studies of poverty and its complex, interrelated issues.
In addition, some statistics can be distorted by bias. For example, Michael Gove recently introduced the baccalaureate in response to inaccurate figures: schools were boasting high pass rates on the basis of vocational subjects, neglecting the poor results in traditional/academic subjects. Such discrepancies with official statistics can invalidate research conducted by governments/schools/sociologists; ultimately, this research forms the basis of sociologist's investigation and argument – sourcing inaccurate information would discredit and undermine the working hypothesis and, consequently, the results would be invalid.
In conclusion, I believe official statistics can be very useful when investigating the effects of material deprivation on educational achievement. In my opinion, they must be properly used in conjunction with qualitative data collected through unstructured interviews and covert participant observation. Official statistics are useful in lending the study balance and a wider sense of scale, but must be complemented with reliable, accurate personal (and detailed) data collected meticulously/properly in order to offer the study all of their benefits and offset their drawbacks. In addition, they must be well sourced and recognised as valid/reliable/accurate results, or else the findings of the research could be considerably discredited/invalidated.
