ASSIGNMENT

AF4S11: Accounting Research Quant Assignment



AF4S11: Accounting Research Quant Assignment

Introduction

This report presents a through statistical analysis of a large dataset obtained from a social survey. To analyze the data we have applied several statistical techniques like descriptive analysis and hypothesis testing. In order to represent the data in graphical form, we used histograms (showing the distribution of data) and box plot. The descriptive are presented under tables. All the techniques are performed by using the SPSS statistical software, the reason of choosing it is that it results provided by this software are much more accurate.

Potential Sources of Data

It is important to be able to rely on the information that has been identified, whether printed or electronic. We must exercise critical thinking to evaluate the quality and it must be borne in mind that any information of unknown origin should be excluded a priori. Documents submitted by the government agencies like GSS are initially evaluated and matched for suitable statistical relevance. To assess the quality and relevance of information, we have taken into account the reputation of the department. For the purpose of this analysis, the potential sources include Gallup, Mori, The Statistical Office of the European Union and International Economic datasets.

Source and Nature of Data

The data for this report is collected from a cross panel survey taken from the General Social Survey (GSS) Data Disk file of Year 2008 labeled as 2008 Cross Panel Data. The data file is appended with this report. Overall, there are 2044 respondents, who represent the data set for region of interviewer and income of respondents. For Hypothesis Testing part, there 3559 respondents, marking variables for family income of respondents and race of respondents. Since the data is considerably large, therefore, it enhances the accuracy of the results because the error is spread across a large number of respondents that were included in the GSS Cross Panel Data Study for the year 2008.

In this report, we examine the measure of central tendency and dispersion for the two variables selected from on the data file. Historical evidences show that the income of an individual is significantly linked with the region of the individual. Region of the interviewer is categorized in nine segments that have been included in the analysis. These segments were categorized on the basis of location of the interviewer. These include New England; middle Atlantic, East North Central, West North Central, South Central, East South Central, West South Central, Mountain, and Pacific region. Since income is taken as a continuous variable therefore, no segment has been created for the total income of the respondent.

Variables and their Meanings in the Data Set

Four variables have been included from the data file in this study; the first two relate to individualized data of the respondents, where one variable is measured as a continuous variable, whereas another variable is measured on a nominal scale. These include “respondent income in dollars” as continuous Variable and the “region of the respondent” as nominal ...