The project
Bars Behind Bars is a project developed for the Open Access and Digital Ethics course of the Master’s degree in Digital Humanities and Digital Knowledge of the University of Bologna. The project aims at analysing the impact of different factors in the context of suicide in prisons across Europe.
This website works as a documentation of the process leading from the choice of the datasets along with their ethical, technological, legal, and qualitative analyses, to the visualisation of the results. The documentation as Jupyter Notebook can be downloaded from the button below.
Further information can be found on GitHub.
11
Datasets
20
Countries
5
Years
+A Good Research Starts With Good Research Questions
Money, Government, and Life Conditions
- How does governments expenditure in justice-related sectors correlate with public perception of the justice system and conditions of living in prison?
- Is there a relation between countries' corruption index and public perception of the justice system?
Life Conditions, Prisoners, and Perception of Justice
- What kind of relationship can be captured between public perception of the justice system and the rate of suicide in prisons?
- Does a relation between corruption index and the rate of suicide in prisons exist? Do they go hand in hand?
- Is there a correlation between living conditions in prison and the rate of suicide in prisons?
Datasets
Datasets have been chosen for the relevance of their contents from Eurostat and dataUNODC.
After the downloads, data have been cleaned by deleting irrelevant entries, duplicates, and columns with non-useful charcaterizations.
Thus, data have been mashed up with the aim of creating datasets specific for the purposes of this study, from which Visualisations and results are derived.
SD: Source Dataset
MUD: Mashed-Up Dataset
* All data and metadata provided on UNdata’s website are available free of charge and may be copied freely, duplicated and further distributed provided that UNdata is cited as the reference
. See reference here.
Datasets
SD1 - Mortality in Prison
SD2 - Persons Held by Sentence Status
SD3 - Prisons Capacity and Number of Persons Held
SD4 - Perceived Independence of the Justice System
SD5 - Corruption Perception Index
SD6 - General Government Expenditure by Function
MUD1 - Deaths in Prison and Governments Expenditure in Justice
MUD2 - Public Perception of Justice System and Corruption Rankings
MUD3 - Deaths in Prison and Public Perception of the Justice System
MUD4 - Deaths in Prison and Corruption Rankings
MUD5 - Deaths in Prison, Unsenteced Prisoners and Prisons' Population
Datasets analyses
Data quality has been checked for each dataset with respect to four principles derived by the Italian National Guidelines for the Enhancement of the Public Information Asset.
Completeness
The datum is exhaustive with respect to every expected value and related entity (sources) that contribute to the definition of the procedure.
Accuracy
The datum and its attributes correctly represet the real value of the concept or event it refers to.
Coherence
The datum and its attributes are not contradictory with respect to other data in the context of use by the administration owner.
Timeliness
The datum and its attributes are up-to-date with respect to the procedure they refer to.
| Id | Completeness | Accuracy | Coherence | Timeliness |
|---|---|---|---|---|
| SD1 |  |
 |
|
|
| SD2 | |
|
|
|
| SD3 | |
|
|
|
| SD4 | |
|
|
|
| SD5 | |
|
|
|
| SD6 | |
|
|
|
+Project sustainibility
Bars Behind Bars
This project has been developed for the "Open Access and Digital Ethics" course of the Digital Humanities and Digital Knowledge Master's Degree (University of Bologna, a.y. 2022/2023). Given the purpose of the project, datasets and results will not be neither updated nor actively maintained in the future.
Source Datasets
The source dataset from which Bars Behind Bars gathered the necessary data are maintained and updated by their proprietary intitutions. Therefore, for information about maintainance we invite the user to follow the links provided in the datasets description section.
Visualisations
+Governments expenditure in justice-related sectors and living conditions in prisons
Select a year and look at the relation between different countries' governemnts expenditure in justice-related sectors and living conditions in prisons.
In the scatter-plot, while expenditure is given as the percentage of the GDP spent by a given country in a specific year, living conditions has been computed as the ratio between the number of prisoners in a given country and the official capacity of that country's prisons.
The term "living conditions" has been used as an indicator of overpopulation in prisons with respect to their normal maximum of hostable prisoners.
Governments expenditure in justice and living conditions in prison for the year 2016
+Corruption index and perception of independence of the justice system
Select a country to automatically modify the chart focusing on its values only.
The higher the value inside the country slice, the higher the perceived corruption. Note that, slice dimension for countries do not change on the basis of the corruption index value since they are not expressed as percentages.
On the contrary, values for perceived independence of the justice system are percentages expressing the answers of a portion of population to questionnarie about the topic. Value are: vary bad (VBAD), fairly bad (FBAD), fairly good (FGOOD), very good (VGOOD) and unknown.
Corruption index and perception of independence of the justice system for the year 2016
Corruption index and perception of independence of the justice system for the year 2017
Corruption index and perception of independence of the justice system for the year 2018
Corruption index and perception of independence of the justice system for the year 2019
Corruption index and perception of independence of the justice system for the year 2020
+Public perception of the justice system and suicide rate in prisons
Interact with the map to change the pie-chart focus on different countries.
By selecting a country, the map will show the number of suicides occurred in the country during the selected year.
Pie-charts highlight the declarations of people with respect to how they perceived their national country's justice system in that specific year. Data are in percentages and have been gathered from questionnaires. Values range from very bad to very good.
Public perception of the justice system and suicide rate in prisons for the year 2016
+Suicides in prisons and corruption index
Select the year to focus on and look at the relation between suicide rate and corruption index for each of the selected countries. Play with the bar-chart to custom the view.
Suicide values are absolute counts and therefore bars for the same country have no aim to be directly compared. Insted, their goal is to show the relation the two variables have if comparison is performed among different countries.
Suicides in prisons and corruption index for the year 2016
+Suicides and living conditions in prisons
Select a year and look at the relation between suicide rate and living conditions in prisons for different countries.
Suicide rate is expressed as the absolute value of suicides occurred in prison in a specific year, while, as before, living conditions has been computed as the ratio between the number of prisoners in a given country and the official capacity of that country's prisons.
Suicides and living conditions in prisons for the year 2016
+Results
Is there a relation between a countries' Government's expenditure in justice-related sectors and living conditions in prisons?
Our concluding remarks after analysing the relation between living conditions in prison and governament expenditure in justice-related sectors are based on the computation of the Pearson correlation coefficient with respect to the cited variables. We see a very very weak negative correlation value of -0.14 and a p-value of 0.2 which in purely statistical terms gives us no apparent correlation between these two variables. Therefore, the negative value would in principle imply that higher expenditure by the governament on maintining these public systems could be a factor in better living conditions in prisons. This correlation seems to be confirmed by the very low p-value, which tells us that this result is very unlikely to be produced by chance. However, the magnitude of this relation is very small even though statisticaaly significant. More data may be helful for computing more precise values to draw conclusive claims and further explore this correlation.
Is there a relation between countries' corruption index and public perception of the justice system?
What is clear from the first sight is that countries with a lower corruption index recorded also a positive perception of the justice system. In fact, focusing on the countries whose corruption index is below 25, it is possible to see how negative opinions are generally the minority. Virtuous examples are northern countries such as Finland, Sweden, and Denmark, which register the lower values of corruption during the years, and also a potive perception above 70%.
The other side of the coin presents a slightly heterogeneous landscape. Public perception change a lot among the countries with a higher corruption index and there are also exceptions such as Hungary, where a high corruption index correspond to a solid preference for a positive opinion towards the justice system.
Other countries, however, show how negative opinions commonly grow when the corruption index is higher. In conclusion, it seems that there is a relation between the two variables, but we should also consider that values expressed derive from public opinions. Overall, data show what it is to some extent natural: it seems that countries where corruption is perceived more, also tend to record more negative opinions about the justice system. The contrary holds too.
What kind of relationship can be captured between public perception of the justice system and the rate of suicide in prisons?
The double chart with the heatmap depicts a miscellaneous scenario. In fact, by selecting the countries with a high suicide rate (such as France) is is possible to see that the perception of the justice system is not necessarily negative. On the contrary, countries with a low suicide rate sometimes show a more negative opinion towards the justice system. This is the case, for example, of Balkan countries.
The only realtion holding between these two variables appear to exists for countries a Italy and Spain, where to a "medium-score" for suicide rate correspond a much more negative opinion towards the national justice system. However, given the absence of enough evidences for other countries, we cannot conclude that, overall, there exists a general relation that bond these two variables. For what concerns countries where this relation seems to hold, it should be necessary to perform a more detailed study on the topic before deriving any further conclusion.
Does a relation between corruption index and the rate of suicide in prisons exist?
Given the data and the visualization (barchart with corruption index and rate of suicide), we can state that no particular and clear relation exists between the two variables in the general scenario. In fact, corruption index and rate of suicides appear to be rather independent from one another and no particular trend can be grasped from the visualization in contemplated years.
However, there are still some peculiarities to be considered. On one hand, France seems to be the country with the highest suicide rate in each of the five years, while its corruption index is, on the other hand, not even close to the highests in the chart.
Moreover, some focus should be dedicated to Italy. Even if we cannot directy compare and derive conclusions from the comparison of the corruption index and the suicide rate (since their values belong to different scales), it is possible to observe how the country recorded one of the highest corruption indexes and suicide rates along the years. This correspondence, however, cannot be officially established with this instruments only. In other words, the visualisation shows a relation that, nevertheless, cannot be declared to have a statistical significance.
Is there a relation between suicides in prisons and living conditions inside prisons?
In this case, after gaving computed the Pearson coefficient and the p-value for the involved variables, we found that the the former amounts to 0.26 while the latter a p-value of 0.02. Statistically speaking, we can say with confidence that the obtained coeffient implies a weak positive correlation between the two variables. Also, the p-value is remarkably low thus allowing us for confidently state that the correlation is statistically significant. In other words, there exists a positive correlation between the suicide rates in prisons and living conditions. The only correlation coefficent, however, does not allow us to determine the direction of the causality and the presence of latent variables having effect on this correlation.
License & Metadata
Following FAIR principles, datasets, as well as their metadata, have been published in a machine readable format and following a common practice in European datasets description for a better findability, accessibility, interoperability, and reusability. Given the nature of the data and their European coverage, it has been chosen to employ the common European data model DCAT_AP 2.0.0 integrated by the use of PROV-O for specifying the derivation of the entities (datasets) this project deals with.
We provided metadata for the catalogue of all the individual datasets derived from the cleaning process, for the mash-up dataset and for the entire cataloge. Each dataset, the catalogue, and the website as well have been published under a Creative Commons license (CC-BY 4.0) following European Commission recommendations and best practices for the re-use of content and data.
+Metadata tables and RDF assertions
Catalogue
About
Bars Behind Bars' purpose if of academic nature only. The project has been developed for the final examination of the Open Access and Digital Ethics course of the Master's Degree in Digital Humanities and Digital Knowledge (University of Bologna).
