Goldsmiths College, University of London
Applied Quantitative Economics
Project
** You must attempt only one project, and you must complete it either in R or in Excel **
General Background
Key Stage 4 (KS4) is a legal term for the last two years of secondary school education in England leading to GCSEs and other examinations, for pupils generally aged between 14 and 16:
Every year, the UK’s Department for Education (DfE) publishes the educational attainment of pupils at the Key Stage 4 (KS4) level and how they compare with other schools in England (click here to access). An accompanying report describes general trends that may be inferred from school-level data (click here to access).
The DfE renders available a full dataset with rich complementary information on pupil characteristics and school expenditures, besides student achievements in GCSE level exams (specific queries to the dataset can be requested from here). However, this complementary dataset is only available for maintained schools, which are government-funded schools that are run by a local authority. Besides maintained schools, the constellation of secondary education institutions in England consists mostly of: academies (government funded but run by an academy trust rather than a local authority), independent schools (privately funded institutions) and special schools (schools that specialise in educating pupils with special educational needs). The information contained in the revised dataset allows us to study different aspects of secondary schools in England.
Because school education up to the age of 16 can be a crucial determinant of career paths, understanding regional inequalities and factors associated with higher student attainment or specific school expenditure patterns can be of outmost importance when designing educational policies. Thus, the aim of this assessment is that you choose and undertake one of the following projects specified below, which must be completed either in R or in Excel.
Project 1: Regional inequalities and determinants of spending by secondary schools in England
School expenditure consists of several budget categories (teaching and support staff, catering, building and energy costs, learning resources and so on) and may be related, amongst other factors, to school income (both funding and self-generated), size, the extent of personalised teaching and its location.
The current state of affairs suggests that local public authorities in England face increasing pressures for constraining expenditure in areas related to schooling, health and social policy. Restrictive policies may trigger a process of fragmentation of social cohesion in the country. Therefore, understanding the overall national evolution and uneven regional dimension of school expenditure, as well as the key determinants of expenditure patterns might aid the design of policies that prevent this fragmentation process from unfolding (or deepening further).
With these considerations in mind, the aim of Project 1 is to study and answer the following three questions:
- What is the extent of regional inequalities in expenditure by secondary schools?
- What has been the recent evolution of total gross expenditure by secondary schools?
- What are the determinants of secondary school expenditure?
Project 2: Regional inequalities and determinants of secondary school student performance in England
Key stage 4 student performance involves attainment in several subjects and has been measured using different indicators. In particular, the Attainment 8 Score measures the achievement of a pupil across 8 qualifications including mathematics and English (each carrying a double weight) and 6 further subjects (an explanation to which can be found here).
Differences in student performance at this stage may be relevant in explaining chances pupils could have in pursuing further studies and a profession. This is even more important when we consider the high degree of specialisation of pupils in the country at an age between 14 and 16.
Even though key crucial determinants of student achievement in exams at this age are beyond the realm of quantitative social research, it is possible to hypothesise that some measurable economic factors like poverty and inequality might play a role in explaining differences in average student performance.
With these considerations in mind, the aim of Project 2 is to study and answer the following three questions:
- What is the extent of regional inequalities in average student performance across secondary schools?
- What has been the recent evolution of income inequality within local authorities where schools with highest and lowest student performance are located?
- What are the factors associated with higher average student performance?
Datasets to be used in both projects
In order to carry out your analysis you will use a large sample from DfE’s dataset ‘Revised key stage 4 results’ for the (academic and fiscal) year 2016-2017, including general school information, student performance and school workforce and finance for hundreds of publicly maintained schools in England. The sample will include the following variables, for each school:
The full sample that you need to use in your project is uploaded on our Moodle VLE page: “AppQE project sample – Full Sample (758 schools).xlsx”. This is the Excel file required to answer questions (i) and question (iii) in each project.
In order to answer question (ii), students who choose Project 01 will have a time series of total gross expenditure and funding of secondary schools in England at the national level between 1999/00 and 2016/17, as well as an economy-wide price index for the corresponding years. File “AppQE TS project 01.xlsx” on our Moodle VLE page contains the data that you will need to answer question (ii).
To answer question (ii), students who choose Project 02 will have a time series of the 80/20 ratio for gross weekly pay in each local authority where the schools of our sample are located covering the 2008-2017 period. File “AppQE TS project 02.xlsx” on our Moodle VLE page contains the data that you will need to answer question (ii).
Project Structure
Your project should be structured along the following lines:
Introduction: motivates the topic under study and lays out the research questions. Its language should be non-technical, accessible to non-specialists. It should also include some economic considerations which motivate you analysis, as well as references to the existing literature;
Dataset characteristics: describes the dataset to be used, acknowledging sources, detailing time-span, geographical scope, units of analysis and variables at your disposal;
Regional inequalities: this section deals with question (i) in each project. Cross- tabulations (pivot and contingency tables), frequency distributions, measures of location (e.g. mean, median, group averages) and dispersion (e.g. standard deviation, inter-quartile range, 80/20 ratio, top 1% percentile) may be used to identify concentration tendencies and characterise regional differences.
Evolution through time: this section deals with question (ii) in each project. Time series techniques may be used to depict and numerically summarise the evolution of schools’ expenditure and compare it to funding provided in England (Project 01), as well as to compare the evolution of the 80/20 ratio of gross weekly pay in selected local authorities in England (Project 02).
Regression analysis: this section deals with question (iii) in each project. Linear regression techniques should be applied to postulate, estimate and test relationships between variables in the dataset, in a way conducive to answer the question posed. The analysis should develop an economic rationale for including the chosen regressors to explain variation in the dependent variable. Point estimates should be reported, statistical significance of coefficients assessed and hypotheses tests carried out. Statistical analysis should be accompanied by an economic analysis of the results, i.e. are statistically significant coefficients coherent with economic intuition and quantitatively relevant? What do the results imply in terms of the economic arguments put forward? Moreover, statistical aspects of this section may be: comparison of actual and fitted values, standardisation of regression variables to compare the magnitude of effects for different regressors, implied trade-off relationships from the estimated model. All results should be reported in an organised way and following standard layouts.
Policy discussion: discusses the implications of the findings for the conduct of policy. This section should connect empirical results with the research questions posed in the Introduction, in view of providing evidence-based suggestions for policy-making.
Concluding remarks: briefly summarise the issues addressed in the project, its most important empirical findings, and policy suggestions.
Word Count
The project has a limit of 4,000 words. The suggested distribution of this figure across the suggested structure is as follows: Introduction, Policy discussion and Concluding remarks (1,000 words); Dataset characteristics, Question (i) and Question (ii) (1,500 words); Question (iii) (1,500 words).
The word limit does not include the R code, in case you choose to complete your project in R. The R code should be embedded within the text but it does not count toward the word limit. If you complete your project in Excel instead, your will submit two files: your project in PDF format and your Excel file separately.
Procedure
- Choose either Project 1 or Project 2
- Choose to complete your project either in Excel or in R.
- Complete your project following the guidelines set in this document.
- If you complete your project in Excel, you will need to submit your project in PDF format and also your Excel file separately. The Excel file will contain all your steps alongside the descriptive statistics, charts, estimation outputs, etc., which have also been included in the project. In the project PDF file you will include only the outputs, not the steps.
- If you complete your project in R, you will submit your entire project in PDF format. The R code must be embedded within the text and pasted just before the outputs, which can be test statistics, tables or plot, similarly to this.
- Submit your project online using the link on our Moodle page.
- Your project will automatically be scanned for plagiarism, and a similarity index will be generated by Moodle. Do not copy the project from someone else as both of you will be penalised. And always cite all your sources. Better safe than sorrow.
- Late submissions will receive a grade penalty unless your extenuating circumstances (EC) claim is approved by the IMS office.
- On our Moodle page you can find two templates for your project. One template uses MS Word and the second template uses Rmarkdown. You are free to use either one of them.
***
Get solved or fresh solution on Applied Quantitative Economics and many more. 24X7 help, plag free solution. Order online now!
