Journal of the Royal Statistical Society: Series A (Statistics in Society)

Cover image for Vol. 179 Issue 3

Edited By: H. Goldstein and L. Sharples

Impact Factor: 1.643

ISI Journal Citation Reports © Ranking: 2014: 11/46 (Social Sciences Mathematical Methods); 19/122 (Statistics & Probability)

Online ISSN: 1467-985X

Associated Title(s): Journal of the Royal Statistical Society: Series B (Statistical Methodology), Journal of the Royal Statistical Society: Series C (Applied Statistics), Significance

177:1


What do healthcare workers know about sudden infant death syndrome?: the results of the Italian campaign ‘GenitoriPiù’, by F. de Luca and G. Boccuzzo, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 63 - 82

Dataset: de Luca - Boccuzzo, Data
Do-file: de Luca - Boccuzzo, Do file

The dataset includes all the variables that were used in the abovementioned paper.
The data belong to the Italian National Campaign ‘GenitoriPiù,’ which was aimed at promoting simple actions proven effective for the prevention of major childhood risks.
The prevention of SIDS was one of the seven objectives of the campaign.
Before the beginning of a training campaign directed at healthcare workers, a survey concerning their attitudes and knowledge was performed between September 2008 and June 2009.
Eleven Italian regions (Abruzzo, Aosta Valley, Apulia, Calabria, Emilia Romagna, Friuli Venezia Giulia, Lazio, Molise, Sardinia, Umbria, and Veneto) and 2 Milan Local Health Units (known in Italian as ASLs) participated in the survey.
A total of 5,911 questionnaires were collected.
The survey cannot be considered representative of the Italian population of healthcare workers, since it is based only on data collected from healthcare workers belonging to the participating regions.
Included in the dataset are the variables regarding the knowledge possessed by healthcare workers over protective factors against SIDS.
The questions required healthcare workers to indicate the effect of the following factors with respect to the degree of protection they provided against SIDS (respondents could choose between ‘protects,’ ‘does not protect,’ and ‘I do not know,’ and the correct answer is here given in brackets):
c1a) Put the newborn to sleep in a supine position [protects];
c1b) Avoid smoking in the room where the newborn sleeps [protects];
c1c) Use a soft mattress for the cot of the newborn [does not protect];
c1d) Breastfeeding [protects];
c1e) Keep high the temperature of the room where the newborn sleeps
c1f) Ensure that the newborn touches the bottom of the cot with her/his feet [protects];
c2) Performing an ECG screening to the newborn [does not protect].
The background variables considered for the healthcare workers included: region to which they belong, gender, age, years of professional experience, professional role and workplace.
To replicate the analysis described in the paper just run the included do-file in STATA.

Federico de Luca
Division of Social Statistics & Demography
School of Social Sciences
University of Southampton
Highfield Campus
Southampton
SO17 1BJ
UK

E-mail: F.Deluca@soton.ac.uk

Dataset

The UK minimum wage at 22 years of age: a regression discontinuity approach, by R. Dickens, R. Riley and D. Wilkinson, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 95 - 114

INSTRUCTIONS FOR CREATING TABLES AND FIGURES

DATA SOURCES:
Access to the data sources used in this paper is restricted for confidentiality reasons. Data sources used are the Quarterly Labour Force Survey (LFS) micro data and the Annual Survey of Hours and Earnings (ASHE) micro data. To arrange access to these please contact:
Secure Data Service, UK Data Archive, University of Essex, Wivenhoe Park, Colchester, Essex CO4 3SQ; Tel: +44 (0)1206 874968; Email: securedata@data-archive.ac.uk; http://securedata.data-archive.ac.uk/.

CREATING THE LFS DATA FILES:
Run syntax files in the order shown

datapost.doRetrieves key variables from the Quarterly Labour Force Surveys from the post NMW period and saves data file "twenty.dta".
dataposta.doUses data file "twenty.dta", creates variables and saves data file "twentya.dta"; this is the main data file used to generate tables and figures.
datapre.doRetrieves key variables from the Quarterly Labour Force Surveys from the pre NMW period and saves data file "twentyp.dta".
dataprea.doUses data file "twentyp.dta", creates variables and saves data file "twentypa.dta".

.

SYNTAX FOR FIGURES 1 AND 2:
Figure 1 Percent of Workers paid below Adult Rate: Labour Force Survey
Figure 2 Employment Rate by Age in Weeks for All Low Skilled Individuals

figs.do Generates the data underlying figures 1 and 2.

SYNTAX FOR TABLE 2:
Table 2 Covariates discontinuities at age 2

22post_quals.doGenerates discontinuity estimates for "no qualifications" age 22 post-NMW, panel 1 Table 2.
22post_white.doGenerates discontinuity estimates for "white" age 22 post-NMW, panel 2 Table 2.
22post_mar.doGenerates discontinuity estimates for "married" age 22 post-NMW, panel 3 Table 2.
22post_sex.doGenerates discontinuity estimates for "sex" age 22 post-NMW, panel 4 Table 2.

.

SYNTAX FOR TABLE 3:
Table 3 Pay discontinuities for the low skilled at age 22, Labour Force Survey Data

22post_lowp.do Generates discontinuity estimates in low pay age 22 post NMW presented in Table 3.

SYNTAX FOR TABLE 4:
Table 4 Pay discontinuities below the adult rate for the low skilled at age 22, Annual Survey of Hours and Earnings
ashe table 4.do Generates ASHE data and estimates regressions in Table 4

SYNTAX FOR TABLE 5:
Table 5 Employment outcomes for the low skilled at age

22post.doGenerates discontinuity estimates presented in Table 5.
22post_spec.doGenerates the fully saturated model with dummy variables for each age measured in weeks. The log-likelihoods from this model are necessary to construct the Chi-squared test statistics in Table 5.

.

SYNTAX FOR TABLE 6:
Table 6 Robustness Checks and Different Outcomes

22pre.doGenerates discontinuity estimates age 22 pre-NMW, panel 1 Table 6.
21post.doGenerates discontinuity estimates age 21 post-NMW, panel 2 Table 6.
23post.doGenerates discontinuity estimates age 23 post-NMW, panel 3 Table 6.
unemp.doGenerates discontinuity estimates for unemployment at age 22 post-NMW, panel 4 Table 6.
ina.doGenerates discontinuity estimates for inactivity at age 22 post-NMW, panel 5 Table 6.

.

Syntax files: 22pre_spec.do, 21post_spec.do, 23post_spec.do, unemp_spec.do, and ina_spec.do generate the fully saturated models with dummy variables for each age measured in weeks. The log-likelihoods from these models can be used to construct the Chi-squared test statistics in Table 6.

SYNTAX FOR TABLE 7:
Table 7 Non-parametric estimates of the discontinuity at age 22

nonpara_22postplus.do Generates non-parametric discontinuity estimates (employment, unemployment and inactivity) presented in Table 7.

SYNTAX FOR TABLE 8:
Table 8 Non-parametric estimates of the discontinuity at ages 21 and 23 and age 22 before the NMW

nonpara_22pre.doGenerates non-parametric discontinuity estimates age 22 pre-NMW, panel 1 Table 8.
nonpara_21post.doGenerates non-parametric discontinuity estimates age 21 post-NMW, panel 2 Table 8.
nonpara_23post.doGenerates non-parametric discontinuity estimates age 23 post-NMW, panel 3 Table 8.

.

Richard Dickens
Department of Economics
University of Sussex
Brighton
BN1 9SL
UK

E-mail: r.f.dickens@sussex.ac.uk

Dataset

A non-parametric model of residual brand equity in hierarchical branding structures with application to US beer data, by S. Voleti and P. Ghosh, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 135 - 152

Programs included:

1. "MDP code in WinBUGS.txt" - This is a WinBUGS program on the mixture of Dirichlet processes (MDP) estimation of the one-sided residual equities at the brand, sub-brand and SKU layers in the branding hierarchy.

2. "SFA code in WinBUGS.txt" - This is the WinBUGS program used to obtain stochastic frontier analysis (SFA) estimates for the benchmark parametric model for the residual equities.

Sudhir Voleti
Indian School of Business
Gachibowli
Hyderabad
India

E-mail: Sudhir_Voleti@isb.edu

Dataset

A comparison of the accuracy of liquid cytology versus conventional screening: a meta-analysis of split-sample studies, by D. Epstein, A. Olry de Labry Lima, L. García Mochón, J. Espín Balbino and Javier Esquivias, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 153 - 168

File 'Test' contains the code and data for the conventional and liquid cytology tests in the paper.

David Epstein
Facultad de Ciencias Economicas
Campus Universitario de Cartuja
Universidad de Granada
Granada 18071
Spain

E-mail: david.epstein@york.ac.uk

Dataset

Form or function?: the effect of new sports stadia on property prices in London, by G. M. Ahlfeldt and G. Kavetsos, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 169 - 190

NOTES:
THE PAPER USES A CONFIDENTIAL DATA SET OF PROPERTY TRANSACTIONS PROVIDED BY THE NATIONWIDE BUILDING SOCIETY, WHICH WE ARE NOT ALLOWED TO SHARE WITH THE WIDER PUBLIC DOMAIN. WE SHARE, HOWEVER, THE COMPUTER CODE USED TO OBTAIN ALL RESULTS PRESENTED * IN THE ARTICLE. THE CODE DEMONSTRATES HOW WE HAVE COMPUTED THE TREATMENT VARIABLES, THE MODELS WE HAVE RUN, AND HOW THE RESULTS HAVE BEEN PROCESSED TO BE ILLUSTRATED IN TABLES AND FIGURES

INSTRUCTIONS:
FOPUB_META IS A META-DO-FILE, WHICH EXPLAINS THE SEQUENCE IN WHICH WE HAVE RUN THREE THREE MORE SPECIFIC DO-FILES

1) FORPUB_WEMB_GRAD: CONTAINS THE DATA PROCESSING AND ESTIMATION RELATED TO NEW WEMBLEY GRADIENT MODELS
2) FORPUB_WEMB_GRID: CONTAINS THE DATA PROCESSING AND ESTIMATION RELATED TO NEW WEMBLEY GRID CELL MODELS
3) FORPUB_WEMB_ARS: CONTAINS THE DATA PROCESSING AND ESTIMATION RELATED TO THE RELOCATION OF THE ARSENAL HOME VENUE

GABRIEL M. AHLFELDT
DEPARTMENT OF GEOGRAPHY AND ENVIRONMENT
SPATIAL ECONOMICS RESEARCH CENTRE
LONDON SCHOOL OF ECONOMICS AND POLITICAL SCIENCE
HOUGHTON STREET
WC2A 2AE
UK

E-mail: G.AHLFELDT@LSE.AC.UK

Dataset

Bayesian ranking responses in multiple-response questions, by H. Wang and W. H. Huang, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 191 - 208

R code
The marginal posterior probability by normal approximation code is in the file MPP.r.

Data
These data are a survey of 49609 first-year college students in Taiwan for the year 2003 about their preferences for college study.

The following question is the multiple response question used in this study.

Question : What kind of experience do you expect to receive during the period of college study? (Select at least one response)

Read Chinese and foreign classics
Travel around Taiwan
Present academic papers in conferences
Lead large-scale activities
Be on a school team
Be a student association leader
Participate internship programs
Fall in love
Have sexual experience
Travel around the world Make many friends
Other

Data file : data_3_Q7
There are 12 responses in the data set.
Response variable :

"1" denotes the respondent select the response ;
"0" denotes the respondent does not select the response ;
Blank denotes the missing value.

Hsiuying Wang
Institute of Statistics
National Chiao Tung University
Hsinchu
30010
Taiwan

E-mail: wang@stat.nctu.edu.tw

Dataset

Multilevel factor analytic models for assessing the relationship between nurse-reported adverse events and patient safety, by L. Diya, B. Li, K. Van den Heede, W. Sermeus and E. Lesaffre, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 237 - 257

Mplus code for EFA
The code in file 'EFA' is the Mplus model code when the cut-off of 2 is used to discretize the data. The first code is for a single-level EFA and the second code is for the multilevel EFA. All EFA models were fitted on the “learning data set”.

R code for single-level CFA
The R code in file 'CFA' is used to fit a single-level CFA in JAGS and to compute the MANOVA and ANOVA discrepancy measures. JAGS is used to estimate the model parameters and to compute the Sums of Squares and Cross Products. The MANOVA and ANOVA discrepancy measures I computed during the post-processing in R. The model fitting is done in JAGS using the latent variable approach (see JAGS model code, 'model.txt') and the discrepancy measures and associated PPP-values are computed using the R function mhier2(obj), where obj is a JAGS object. The “validation” data set is used to fit the single-level CFA and model assessment is done on the “learning data set”. In this analysis we used the total data set and the variable samp is an indicator for the validation sample (samp=1; 0 otherwise). The code can be easily extended to cater for multilevel CFA.

R function to compute MANOVA and ANOVA discrepancy measures
The file 'mhier2(obj)' contains the code to compute the MANOVA and ANOVA discrepancy measures.

Luwis Diya
Interuniversity Institute for Biostatistics and Statistical Bioinformatics
KU Leuven
Blok D
Bus 7001
Kapucijnenvoer 35
B3000 Leuven
Belgium

E-mail: luwis.diya@yahoo.co.uk

Dataset

Treatment comparisons for decision making: facing the problems of sparse and few data, by M. O. Soares, J. C. Dumville, A. E. Ades and N. J. Welton, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 259 - 279

File 'Code and data' contains the Winbugs code and data for models A1 and B1 in the paper.

Marta O. Soares
Centre for Health Economics
Alcuin A Block
University of York
Heslington
York
YO10 5DD
UK

E-mail: marta.soares@york.ac.uk

Dataset

Synthesis of evidence on heterogeneous interventions with multiple outcomes recorded over multiple follow-up times reported inconsistently: a smoking cessation case-study, by J. Madan, Y.-F. Chen, P. Aveyard, D. Wang, I. Yahaya, M. Munafo, L. Bauld and N. Welton, Journal of the Royal Statistical Society, Series A, Statistics in Society, Volume 177, part 1 (2014), pages 295 - 314

The file 'Winbugs NWMA' contains the winBUGS code for model A3.3.1 with classlevel treatment effects, Weibull time-to-relapse, and a bivariate normal distribution relating intervention effects on the two outcome measures (with intercept assumed to be zero). Code for other models can be derived from this version with minimal changes (details are available from the corresponding author on request).

Jason Madan
Warwick Clinical Trials Unit
Division of Health Sciences
Warwick Medical School
University of Warwick
Coventry
CV4 7AL
UK

E-mail: J.J.Madan@warwick.ac.uk

Dataset

SEARCH

SEARCH BY CITATION