Further information Statistical Training Unit Administrator Room 1127 Government Buildings Cardiff Road Newport NP10 8XG Email: [email protected]1 Statistical Training Unit - Prospectus 2014-15 GSS Statistical Training Programme September 2014 - July 2015 GSS Stat training program Oct 2014.indd Spread 1 of 12 - Pages(24, 1) GSS Stat training program Oct 2014.indd Spread 1 of 12 - Pages(24, 1) 07/10/2014 10:46 07/10/2014 10:46
12
Embed
GSS Statistical Training Programme · 2016. 6. 21. · Geography 15 Sample Design and Estimation 15,16 ANOVA 17 Simple Linear Regression 18 Multiple Linear Regression 18 Introduction
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Further informationStatistical Training Unit AdministratorRoom 1127Government BuildingsCardiff RoadNewportNP10 8XG
GSS Stat training program Oct 2014.indd Spread 1 of 12 - Pages(24, 1)GSS Stat training program Oct 2014.indd Spread 1 of 12 - Pages(24, 1) 07/10/2014 10:4607/10/2014 10:46
2 Statistical Training Unit - Prospectus 2014-15
WelcomeThe GSS Statistical Training Programme is a series of short courses offered by experienced methodological experts from the Offi ce for National Statistics (ONS).
Aim of the programmeThe GSS Statistical Training Programme delivers training on the main methods used in Offi cial Statistics. The courses provide a useful overview of the statistical techniques used by statisticians and provide information on how and why to choose the most appropriate methodology. All courses are delivered by experts within the fi eld and are coordinated by the Statistical Training Unit (STU) at the ONS.
STU aims to provide a wide selection of courses that meet the needs of staff who are involved with the collection, analysis or dissemination of offi cial statistics. The aim is to provide statistical courses that strengthen and/or refresh professional skills and knowledge.
ContentsQuality and Statistics 4
Introduction to National Acounts 5
Administrative Data 6
Record Linkage 6
Introduction to Questionnaire Design and Testing 7
Statistical Disclosure Control 7
Editing and Imputation 8
Digital Dissemination 10
Data Visualisation 10
Randomised Control Trials 11
Hypothesis Testing 11
Index Numbers 12
Seasonal AdjustmentX-13-ARIMA-SEATS 14
Small Area Estimation 14
Geography 15
Sample Design and Estimation 15,16
ANOVA 17
Simple Linear Regression 18
Multiple Linear Regression 18
Introduction to Logistic Regression 19Generalised Linear Models 19
RSS Ordinary Certifi cate 20,21
Course Timetable Information 23
23Statistical Training Unit - Prospectus 2014-15
The Statistical Training Unit (STU) is pleased to offer the GSS Statistical Training Programme as part of Continuing Professional Development (CPD). The courses aim to deliver training in the main methods used in offi cial statistics. The courses provide a useful overview of the statistical techniques used by statisticians and provide information on how and why to choose the most appropriate methodology.
All courses are delivered by experienced methodological experts and are delivered at various ONS sites: Newport, Titchfi eld and London.
STU aims to to provide a wide selection of courses that meet the needs of staff who are involved with the collection, analysis or dissemination of offi cial statistics. The aim is to provide statistical courses that strengthen and refresh professional skills and knowledge.
GSS Statistical Training Programme Course timetable information 2014/15
For more information and course timetables please visit website:https://gss.civilservice.gov.uk/people-and-careers/courses-and-events/
statistical-training-unit/
GSS Stat training program Oct 2014.indd Spread 2 of 12 - Pages(2, 23)GSS Stat training program Oct 2014.indd Spread 2 of 12 - Pages(2, 23) 07/10/2014 10:4607/10/2014 10:46
22 Statistical Training Unit - Prospectus 2014-15 3Statistical Training Unit - Prospectus 2014-15
How to apply• Complete the course applicaton form shown on the GSS Civil
Service Website and email to• [email protected], or• Telephone 01633 455627
GSS Stat training program Oct 2014.indd Spread 3 of 12 - Pages(22, 3)GSS Stat training program Oct 2014.indd Spread 3 of 12 - Pages(22, 3) 07/10/2014 10:4607/10/2014 10:46
4 Statistical Training Unit – Prospectus 2014-15
Quality and Statistics – Level 1
Aims
This course aims to introduce some of the quality considerations that are appropriate in the production of offi cial statistics.
Objectives
Through aligning with specifi c aspects of the Code of Practice and sharing recent guidance from the GSS Statistical Policy and Standards Committee, course participants will gain:
• an understanding of the importance of quality,
• knowledge in the practical application of quality and how to implement it, and
• an overview of relevant guidance and tools.
Duration: 1 day
Entry requirements
Provides an introduction and understanding of the subject – no prior knowledge required.
Degree in maths/statistics or a good working knowledge of the subject is pre-requisite of a Level 2 course.
When to apply
There is no formal closing date for applications, but early application is recommended to secure a place on the course.
Costs
0.5 day course £74
1 day course £111
1.5 day course £160
2 day course £197
0.5 day course £79
1 day course £121
1.5 day course £170
2 day course £207
N
L1
L1
L2
LEVEL 1
LEVEL 2
N T
L
T
21Statistical Training Unit - Prospectus 2014-15
Unit 3
Condensing, Presenting and Summarising Data
• Condense, present and summarise raw data
• Condense, present and summarise frequency data
• Condense, present and summarise grouped data
Unit 4
Time Series and Index Numbers
• Drawing a time chart
• The components of a time series
• The most appropriate model to fi t to a time series and how to fi t it
• What an index number is and why it is used
• Differences between price and quantity indices
• How to calculate different indices
Unit 5
Regression and Correlation
• What regression is and why it is used
• Calculating and interpreting the values of the regression line
• How to draw a scatter diagram and describe the correlation between
• the variables
• Interpreting correlation coeffi cients
Unit 6
Probability
• The defi nition of probability
• How to calculate probabilities
• Mutually & non-mutually exclusive events
• Independent & dependent events
• Venn and tree diagrams
• Conditional probabilities
GSS Stat training program Oct 2014.indd Spread 4 of 12 - Pages(4, 21)GSS Stat training program Oct 2014.indd Spread 4 of 12 - Pages(4, 21) 07/10/2014 10:4607/10/2014 10:46
20 Statistical Training Unit - Prospectus 2014-15
RSS Ordinary Certifi cate
Target audience
Staff who wish to develop their skills and understanding of specifi csurvey processes and/or want to obtain a recognised qualifi cation.Applicants should have at least a GCSE in Mathematics.
Course outline
The RSS Ordinary course is delivered between September and May eachyear. Distance learning is available for those unable to attend aweekly class.
Classes take place ot ONS offi ces in Newport and Titchfi eld
For those with more specifi c trainingneeds, one-off units will be available on a ‘fi rst come, fi rst served’ basis.
Research Methods, Data Collection and Questionnaire Design
• Distinction between observational and experimental studies
• The origin, use and interpretation of published or administrative data
• Overview of offi cial statistics
• Different types of surveys
• Design of a simple questionnaire
Unit 2
Sampling Frames, Methods and Errors
• Defi nitions of a population and a sampling frame
• Differences between random and non-random sampling
• The different methods for selecting a sample
• Types of non-sampling errors
5Statistical Training Unit – Prospectus 2014-15
Introduction to National Accounts – Level 1
Aims
The aims of this course are to introduce participants:
• to national accounts terminology,
• the division of the economy into sectors, and
• the different approaches to measuring Gross Domestic Product (GDP).
Objectives
By the end of the course, participants will:
• have a basic understanding of the language of national accounts,
• be able to list the characteristics of the sectors of the UK, and
• describe the three approaches to measuring GDP.
Duration: 1 day
NL1 Locations
The courses will run at the Offi ce for National Statistics in Newport, Titchfi eld and London.Each location is represented by a colour key seen below and throughout the programme.
NEWPORT
TITCHFIELD
LONDON
N
T
L
GSS Stat training program Oct 2014.indd Spread 5 of 12 - Pages(20, 5)GSS Stat training program Oct 2014.indd Spread 5 of 12 - Pages(20, 5) 07/10/2014 10:4607/10/2014 10:46
6 Statistical Training Unit - Prospectus 2014-15
L1
Administrative Data – Level 1
Aims
The aims of the course are to:
• consider the differences between survey and administrative data,
• highlight some of the challenges to be overcome when linking different datasets, and
• provide an insight to the legal issues around acquiring administrative data.
Objectives
By the end of the course, participants will be able to:
• explain what is meant by the term administrative data,
• describe the benefi ts and limitations of using administrative data,
• explain the advantages of using administrative data in the production of social and demographic statistics, and
• describe the principles behind matching and linking microdata.
Duration: 1 day
Record Linkage – Level 1
Aims
The aim of the course is to identify which records in two different sets of data relate to the same individuals (ie matching or record linkage).
Objectives
Through practical application of particular matching methods, participants will be able to understand:
• why record linkage is not straightforward,
• the desirable qualities of matching variables,
• how to make the task manageable with very large datasets,
• the different types of matching available,
• clerical matching and how it fi ts into a wider matching context, and
• striking the balance between fi nding matches and not making too many false matches.
Duration: 1 day
L1 NT T L
19Statistical Training Unit - Prospectus 2014-15
TNL1
Introduction to Logistic Regression
Aims
The aims of this session are to:
• revise the use of logistic regression,
• assess model fi t and interpretation of model parameters using odds ratios, and
• predicted probabilities, marginal effects.
Alternative link functions, for example probit, will be discussed but not in detail.
Objectives
By the end of the session, participants will gain a refresher in the application and interpretation of logistic regression.
Duration: 0.5 day
Generalised Linear Models (GLMs) – Level 1
Aims
The aims of this session are to:
• review the GLM framework
• introduce specifi c models for different outcome variables, including count variables, ordinal variables, and nominal variables, and
• interpret model parameters and assess model fi t.
Objectives
By the end of the session, participants will gain a refresher in the application and interpretation of GLMs.
Duration: 1 day
TNL1 LL
GSS Stat training program Oct 2014.indd Spread 6 of 12 - Pages(6, 19)GSS Stat training program Oct 2014.indd Spread 6 of 12 - Pages(6, 19) 07/10/2014 10:4607/10/2014 10:46
18 Statistical Training Unit - Prospectus 2014-15
Simple Linear Regression
Aims
The aims of this session are to:
• revise the use of simple linear regression,
• assess the fi t of the model,
• test and interpret the estimated model, and
• discuss the use of residual analysis to validate model assumptions.
Transforming the data to improve model fi t will also be introduced but not in detail.
Objectives
By the end of the session, participants will gain a refresher in the application and interpretation of simple linear regression.
Duration: 0.5 day
Multiple Linear Regression
Aims
The aims of this session are to:
• revise approaches to building a multiple regression model, including stepwise procedures, handling categorical variables, non-linear relationships, and interactions,
• select and interpret suitable models, and
• discuss the different roles of modelling.
Objectives
By the end of the session, participants will gain a refresher in the application and interpretation of multiple linear regression.
Duration: 0.5 day
TNL1 TNL1L L
7Statistical Training Unit - Prospectus 2014-15
Introduction to Questionnaire Design And Testing – Level 1
Aims
The course aims to introduce you to data collection methodology and the role of ONS’ Data Collection Methodology branch in the survey development process.
Objectives
The course includes the following topics, covering both basic principles and some more advanced issues:
• sources of non-sampling error, focusing on measurement error and non-response error,
• modes of data collection including mixed modes,
• questionnaire and question design principles, including different types of questions,
• measurement error including mode effects, and potential ways to mitigate them,
• qualitative and quantitative methods for developing and testing questions, including expert review, and
• cognitive question testing.
You will have the opportunity to try out for yourselves some of the work that DCM does through interactive exercises.
Duration: 1 day
Statistical Disclosure Control – Level 1
Aims
The aims of the course are to introduce participants to:
• the potential disclosure risks, and • the statistical disclosure control
methods available.
Objectives
By the end the course, participants will be able to understand:
• what is meant by statistical disclosure risk and statistical disclosure control,
• why there is a need to protect data against disclosure,
• how to recognise situations that bring about statistical disclosure risk,
• the disclosure techniques and tools used to protect data, and
• the GSS standards for disclosure control.
Duration: 1 day
L1 L1N NT TL L
GSS Stat training program Oct 2014.indd Spread 7 of 12 - Pages(18, 7)GSS Stat training program Oct 2014.indd Spread 7 of 12 - Pages(18, 7) 07/10/2014 10:4607/10/2014 10:46
8 Statistical Training Unit - Prospectus 2014-15
N
Editing and Imputation for Business Surveys – Level 1
Aims
The aims of this course are to introduce participants working with business surveys to:
• Editing – the process of detecting and correcting errors in business survey response data, and
• imputation – the process of estimating for non-respondents to business surveys.
Objectives
By the end of the course participants will be able to understand:
• the vital importance of editing and imputation, and
• how editing and imputation methods are implemented in practice.
Duration: 1 day
Editing and Imputation for Business and Social Surveys – Level 2
Aims
The aims of this course are to introduce participants to the range of editing and imputation methods available, and factors relating to their suitability in business and social surveys.
Objectives
Through a mixture of theory and examples, participants will be gain an understanding of:
• the different data capture methods used by ONS, and errors that can arise at data collection,
• the methods to effi ciently identify and correct for errors in business and social surveys,
• different levels of non-response, and how to deal with missing values, and
• the Editing & Imputation strategy implemented by ONS, and the reasons for the methods applied.
Duration: 1 day
L1 N L2
17Statistical Training Unit - Prospectus 2014-15
ANOVA
Aims
The aims of this session are to:
• revise the use of ANOVA and explore differences in means across factors,
• consider the key assumptions behind ANOVA,
• interpret basic ANOVA output, and
• discuss additional analyses that can help the understanding of results.
Extensions to ANCOVA and MANOVA are introduced but not in detail.
Objectives
By the end of the session, participants will gain a refresher in the application and interpretation of ANOVA.
Duration: 0.5 day
LTNL1
GSS Stat training program Oct 2014.indd Spread 8 of 12 - Pages(8, 17)GSS Stat training program Oct 2014.indd Spread 8 of 12 - Pages(8, 17) 07/10/2014 10:4607/10/2014 10:46
16 Statistical Training Unit - Prospectus 2014-15
Sample Design and Estimation for Social Surveys – Level 2
Aims
The aim is for participants to explore aspects of sampling and estimation that are particularly applicable to social (household) surveys. A parallel course is available for business surveys.
Objectives
Through practical application of the methods participants will understand:
• the principles and practicalities behind sampling and estimation used within offi cial statistics,
• when to use different sampling strategies, with particular emphasis on multi-stage sampling,
• how sample data can be used to estimate parameters of interest, with emphasis on weighting strategies that account for non-response and allow calibration to known population totals, and
• how to assess the quality of sample estimates, with attention paid to sampling errors and the use of design factors.
Duration: 1 day
Sample Design and Estimation for Business Surveys – Level 2Aims
The aim is for participants to explore aspects of sampling and estimation that are particularly applicable to business (or establishment) surveys. A parallel course is available for social surveys.
Objectives
Through practical application of the methods participants will understand:
• the principles and concepts behind sampling and estimation used within offi cial statistics,
• when to use different sampling strategies, with particular emphasis on stratifi cation,
• how sample data can be used to estimate parameters of interest, with emphasis on estimation methods that make best use of other available information, and
• how to assess the quality of sample estimates, including the impact of the sample design and estimator.
Duration: 1 day
NL2 NL2L
9Statistical Training Unit - Prospectus 2014-15
GSS Stat training program Oct 2014.indd Spread 9 of 12 - Pages(16, 9)GSS Stat training program Oct 2014.indd Spread 9 of 12 - Pages(16, 9) 07/10/2014 10:4607/10/2014 10:46
10 Statistical Training Unit - Prospectus 2014-15
T
Digital Dissemination – Level 1
Aims
The aims of the course are to introduce participants to:
• the use of the Internet and how we disseminate statistics,
• how we can learn from common practices within the commercial world to help us disseminate better,
• the challenges specifi c to statistical output, and
• the skills needed to take full advantage of the web.
Objectives
By the end of the course, participants will gain an understanding in:
• why, what and how we disseminate, and
• how we will disseminate in the future.
Duration: 1 day
Data Visualisation – Level 1
Aims
The aims of the course are to introduce course participants to:
• some of the problems that people may have in perceiving the information encoded in a table and/or graph, and
• options for overcoming these problems.
Objectives
By the end of course, participants will gain an understanding of the fundamental design strategies needed to overcome the problems with the perception of those reading statistical tables and graphs.
Duration: 1 day
NL1 TNL1L L
15Statistical Training Unit - Prospectus 2014-15
Geography – Level 1
Aims
The aims of the course are to introduce course participants to:
• The importance and use of geography in the production of statistics, and
• The generic elements of geographic knowledge, understanding and skills needed for analysis and research.
Objectives
By the end of the course, participants will gain an understanding of:
• The complexity of statistical geography in the UK,
• Basic spatial statistical techniques,
• Interpreting statistical maps,
• the application of the Geography policy for National Statistics, and
• ONS’s Open Geography portal, resources and tools.
Duration: 1 day
Sample Design and Estimation for Social Surveys – Level 1
Aims
The aim of the course is to introduce participants, who work in social surveys, to the methods used in sample design and estimation.
Objectives
Through practical application of the methods discussed, course participants will be able to understand:
• sources of non-sampling error, focusing on measurement error and nonresponse error,
• modes of data collection including mixed modes,
• questionnaire and question design principles, including different types of questions, sources of measurement error including mode effects, and potential ways to mitigate them, and
• qualitative and quantitative methods for developing and testing questions, including expert review and cognitive question testing.
Duration: 2 days
L1 T L1 N
GSS Stat training program Oct 2014.indd Spread 10 of 12 - Pages(10, 15)GSS Stat training program Oct 2014.indd Spread 10 of 12 - Pages(10, 15) 07/10/2014 10:4607/10/2014 10:46
14 Statistical Training Unit - Prospectus 2014-15
Seasonal Adjustment X-13-ARIMA-SEATS – Level 2
Aims
The aims of this course are to introduce course participants to:
• the theory behind seasonal adjustment and the seasonal adjustment process, and
• the US Census Bureau’s software X-13-ARIMA-SEATS.
Objectives
a mixture of theaory and the practical applicaton of X-13-ARIMA-SEATS, course participants will gain an understanding of:
• the theory underpinning seasonal adjustment,
• how to perform basic seasonal adjustment, and
• how to analyse your own time series data.
Duration: 1.5 days
Small Area Estimation – Level 2
Aims
The aim of the course is to introduce participants to:
• the methodology applied to small area estimates, and
• when and how small area estimation methods should be applied.
Objectives
Through a mixture of theory and practical application, by the end of the course, participants will be able to combine and model survey data with other datasets, eg census/admin data and understand:
• concepts and principles of a variety of methods,
• which methods are most commonly used and why,
• how different methods are applied, and
• techniques used in Offi cial Statistics.
Duration: 1 day
N TL2 L2
11Statistical Training Unit - Prospectus 2014-15
Randomised Control Trials – Level 2
Aims
The aims of the course are to introduce participants to the theory behind Randomised Control Trials and to give an introduction to good practice guidelines for their practical application.
Objectives
Through the use of examples and theory, participants will gain an understanding of:
• what Randomised Control Trials are,
• how they work, and
• the benefi ts and limitations of their application.
Duration: 1 day
Hypothesis Testing – Level 2
Aims
The aim of the course is to give participants an overview of Hypothesis Testing and its application within offi cial statistics.
Objectives
Through the use of examples and theory, course participants will gain:
• an understanding of the basics of hypothesis testing,
• a working knowledge of statistical power,
• a toolkit of appropriate tests and information on when to use them, and
• an overview of more complex hypothesis testing problems and knowledge of how to deal with them.
Duration: 1 day
NNL2 L2T TL
GSS Stat training program Oct 2014.indd Spread 11 of 12 - Pages(14, 11)GSS Stat training program Oct 2014.indd Spread 11 of 12 - Pages(14, 11) 07/10/2014 10:4607/10/2014 10:46
12 Statistical Training Unit - Prospectus 2014-15
Index Numbers – Level 1
Aims
The aim of this workshop is to introduce participants to the theory behind index numbers and the techniques used in practice.
Objectives
By the end of the course, participants will have an understanding of the following topics:
• the uses of index numbers,
• value and value share,
• Laspeyres and Paasche indices,
• the choice of index formula, and
• defl ation and referencing.
Duration: 1 day
Index Numbers – Level 2
Aims
The aims of the course are to introduce participants to:
• The theoretical background to index numbers
• Index number techniques used in practice, and
• Aspects of the Consumer Price Index.
Objectives
By the end the course, participants will be able to understand:
• Value and value share
• Laspeyres and Paasche indices
• Understanding the Laspeyres-Paasche difference
• The choice of index formula
• Domains and aggregation
• Defl ation, referencing and linking,
• Aspects of the Consumer Prices Index, and
• Wider uses of index numbers.
Duration: 1.5 days
NL1 NL2
13Statistical Training Unit - Prospectus 2014-15
GSS Stat training program Oct 2014.indd Spread 12 of 12 - Pages(12, 13)GSS Stat training program Oct 2014.indd Spread 12 of 12 - Pages(12, 13) 07/10/2014 10:4607/10/2014 10:46