This ReadMe file lists and describes the IHM (individual household method) data and relevant documents for the 'Assessing Rural Transformations' (ART) project, under the following headings: 1. 'Instructions for accessing and interpreting IHM data from the ART project.docx' 2. 'open-IHM manuals' 3. 'ART databases anonymised' 4. 'Household lists, with beneficiary information' 5. 'Non-automated input costs' 6. 'ART DI_AEs, HHs included in final analysis' N.B. Data is grouped around four top-level sites: Karonga district, Lilongwe district (sometimes referred to as 'Masumbankhunda'), Oromia (sometimes referred to as 'Assela') and Tigray. The data and other documents are as follows: 1. 'Instructions for accessing and interpreting IHM data from the ART project.docx' This Word document contains instructions for accessing and interpreting the ART project's IHM data, with reference to many of the different documents uploaded: the open-IHM installation and uninstallation guides, the open-IHM user guide, the food energy values, the anonymised open-IHM databases, and the spreadsheets of non-automated input costs. 2. 'open-IHM manuals' This folder contains four documents that may be needed for access to open-IHM data: 2a. 'open-IHM installation guide (July 2014).pdf' This is a guide to installing the (open-source) 'open-IHM' software - a prerequisite for accessing the IHM data. 2b. 'Uninstalling open-IHM.pdf' This guide to uninstalling open-IHM may need to be used if you make any errors in installing open-IHM with the above installation guide (before then trying to reinstall open-IHM - perhaps trying to follow the installation instructions more closely), or if you later wish to remove open-IHM from your computer. 2c. 'open-IHM user guide (September 2013).pdf' This is a general guide* to using open-IHM software - covering all of the main functions for creating and importing databases, adding and editing individual household data and generating output. * N.B. The 'Instructions for accessing and interpreting IHM data from the ART project.docx' Word document contains more specific instructions for generating output from these ART project datasets, and should be your primary reference point for this. However, the user guide may helpfully supplement some of these instructions. 2d. 'Food energy values, July 2015.xls' This spreadsheet of food energy values should be imported into your fresh installation of open-IHM, before producing any output from the software. The 'Instructions for accessing and interpreting IHM data from the ART project.docx' Word document contains instructions for doing this, as well as instructions for making database-specific changes to the food energy values for some of the databases. You should not need to make any changes to this 'Food energy values, July 2015.xls' spreadsheet - however: The 'Name' column refers to the name of the food (or other crop, livestock product or wild food) type. The 'Category' column shows whether it is categorised as a crop, a livestock product or a wild food. The 'EnergyValuePerUnit' column shows the kilocalorie value per unit of the food (or other crop, livestock product or wild food) type - this relates to the 'UnitOf Measure' in the next column. The 'UnitOfMeasure' column shows the unit of measure in which the food (or other crop, livestock product or wild food type) is recorded and to which its kcal value per unit relates. 3. 'ART databases anonymised' This folder contains the final, anonymised baseline and endline ART IHM databases, accessible with (open-source) open-IHM software. The 'Instructions for accessing and interpreting IHM data from the ART project.docx' Word document brings together instructions for installing open-IHM, importing these databases and producing output from them. 3a. 'Assela (Oromia)' This folder contains the baseline and endline ART IHM databases for Oromia, Ethiopia. The baseline 2012 IHM data is spread across two databases: 'Golla Village Anonymised.ihm' and 'Malt Barley Beneficiaries Anonymised.ihm'. For analysis, output from these two databases should be merged/combined (e.g. by copying and pasting the results from one below the results from the other in Excel, and then sorting all results by the appropriate column). The endline 2014 IHM data is in the 'malt barley assela round 3 anonymised.ihm' database. 3b. 'Masumbankhunda' This folder contains the baseline and endline ART IHM databases for Lilongwe district, Malawi. The baseline 2012 IHM data is in the 'Masumbankhunda 2012 anonymised.ihm' database. The endline 2014 IHM data is in the 'Masumbankhunda 2014 Anonymised.ihm' database. 3c. 'Maulunga-Karonga (Karonga)' and 'Mkango-Karonga (Karonga)' These folders contain the baseline and endline ART IHM databases for Karonga district, Malawi, for each year spread across two databases for separate villages. For most analysis, output from the two databases for each year should be merged/combined (e.g. by copying and pasting the results from one below the results from the other in Excel, and then sorting all results by the appropriate column) - as part of this, in other documents here the households from Maulunga have had 80,000 added to their household ID numbers/codes, and the households from Mkango have had 90,000 added to their household ID numbers/codes (to avoid any confusing duplication). The baseline 2013 IHM data is in the 'Maulunga 2013 Anonymised.ihm' and 'Mkango 2013 Anonymised.ihm' databases. The endline 2015 IHM data is in the 'Maulunga 2015 Anonymised.ihm' and 'Mkango 2015 Anonymised.ihm' databases. 3d. 'Tigray' This folder contains the baseline and endline ART IHM databases for Tigray, Ethiopia. The baseline 2014 IHM data is in the 'Tigray 2014 Anonymised.ihm' database. The endline 2015 IHM data is in the 'Tigray 2015 Anonymised.ihm' database. 4. 'Household lists, with beneficiary information' This folder contains the final/fully up-to-date beneficiary statuses for all households included in the final IHM analysis for the ART project, together with the identifying IHM 'household numbers' of all these households and the codes used to identify these households in QUIP studies (where relevant). Household who were not present in both years, or for whom there were outstanding queries are excluded from these lists. 4a. 'Karonga IHM list of households included in final analysis.xlsx' This spreadsheet contains information on the households included for the final Karonga district IHM analysis. In this spreadsheet: The 'IHM household number' column shows the final IHM identifying household number/ code of each included household. The 'QUIP code' column shows the code used to identify the household in QUIP studies, where the household was also included in the QUIP studies - with blanks for households not included in the QUIP studies. The 'Detailed beneficiary status' column shows whether households were beneficiaries or non-beneficiaries in the years between the baseline and endline; for beneficiaries the information here includes the crops for which inputs were received and the year(s) of this. The 'Simpler beneficiary status' column shows a simplified version of the beneficiary status from the previous column, with fewer different categories. The 'Village' column shows whether the household is in Maulunga or Mkango village. 4b. 'Masumbankhunda IHM list of households included in final analysis.xlsx' This spreadsheet contains information on the households included for the final Lilongwe district IHM analysis. In this spreadsheet: The 'IHM household number' column shows the final IHM identifying household number/ code of each included household. The 'QUIP code' column shows the code used to identify the household in QUIP studies, where the household was also included in the QUIP studies - with blanks for households not included in the QUIP studies. The 'Detailed beneficiary status' column shows whether households were beneficiaries or non-beneficiaries in the years between the baseline and endline; for beneficiaries the information here includes the crops for which inputs were received and the year(s) of this. The 'Most simple beneficiary status' column shows a simplified version of the beneficiary status from the previous column, with fewer different categories. 4c. 'Oromia ('Assela') IHM list of households included in final analysis.xlsx' This spreadsheet contains information on the households included for the final Oromia IHM analysis. In this spreadsheet: The 'IHM household number' column shows the final IHM identifying household number/ code of each included household. The 'QUIP code' column shows the code used to identify the household in QUIP studies, where the household was also included in the QUIP studies - with blanks for households not included in the QUIP studies. The 'Beneficiary status' column shows whether households were beneficiaries or non-beneficiaries in the years between the baseline and endline; for beneficiaries the information here includes the year in which they were beneficiaries. The 'In Golla village?' column shows a 'Y' for households located in Golla village, and a blank for households located elsewhere (but nearby). 4d. 'Tigray IHM list of households included in final analysis.xlsx' This spreadsheet contains information on the households included for the final Tigray IHM analysis. In this spreadsheet: The 'IHM household number' column shows the final IHM identifying household number/ code of each included household. The 'QUIP code' column shows the code used to identify the household in QUIP studies, where the household was also included in the QUIP studies - with blanks for households not included in the QUIP studies. The 'Detailed beneficiary status' column shows whether households were beneficiaries or non-beneficiaries in the years between the baseline and endline; for beneficiaries the information here includes the interventions for which inputs were received and any relevant yearly variation. The 'Simpler beneficiary status' column shows a simplified version of the beneficiary status from the previous column, with fewer different categories. Households may be classified here as 'non-beneficiaries' if they received an intervention in the endline year that was considered unlikely to generate income in the study period. 5. 'Non-automated input costs' This folder contains input costs that have not been automated or already included in other ways within the final ART IHM databases, but which must be subtracted appropriately from some initial open-IHM output on income, disposable income, the standard of living threshold, simulations, and also taken into account for other output where households are ranked by disposable income. These non-automated input costs affect the 2014 Tigray IHM database and both the 2012 and 2014 Oromia databases. The 'Instructions for accessing and interpreting IHM data from the ART project.docx' Word document has some further background and instructions for carrying out these calculations. The instructions are not fully comprehensive, however, as there could be many possible permutations - and so some degree of applying broader principles will be needed here. Contact Celia Petty at Evidence for Development (info@evidencefordevelopment.org) for advice on specific questions. 5a. 'Assela (Oromia) 2012 datasets' This folder contains the non-automated input costs for the baseline ART IHM databases for Oromia, Ethiopia. The 'Assela (Oromia) 2012 'Golla village' non-automated input costs.xlsx' spreadsheet corresponds to the 'Golla Village Anonymised.ihm' baseline open-IHM database, and the 'Assela (Oromia) 2012 'malt barley beneficiaries' non-automated input costs.xlsx' spreadsheet corresponds to the 'Malt Barley Beneficiaries Anonymised.ihm' baseline open-IHM database. In both spreadsheets: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'Agricultural inputs: total cost' column shows each household's total cost ('raw' [i.e. not standardised for household size], and in Ethiopian birr) of purchasing agricultural inputs during the study period. The 'Livestock inputs: total cost' column shows each household's total cost ('raw' [i.e. not standardised for household size], and in Ethiopian birr) of purchasing livestock inputs during the study period. 5b. 'Assela (Oromia) 2014 dataset' This folder contains the non-automated input costs for the endline ART IHM database for Oromia, Ethiopia. The 'Assela (Oromia) 2014 non-automated input costs.xlsx' spreadsheet corresponds to the 'malt barley assela round 3 anonymised.ihm' endline open-IHM database. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'Agricultural inputs: total cost' column shows each household's total cost ('raw' [i.e. not standardised for household size], and in Ethiopian birr) of purchasing agricultural inputs during the study period. The 'Livestock inputs: total cost' column shows each household's total cost ('raw' [i.e. not standardised for household size], and in Ethiopian birr) of purchasing livestock inputs during the study period. 5c. 'Tigray 2014 dataset' This folder contains the non-automated input costs for the baseline ART IHM database for Tigray, Ethiopia. The 'Tigray 2014 non-automated input costs.xlsx' spreadsheet corresponds to the 'Tigray 2014 Anonymised.ihm' baseline open-IHM database. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'Agricultural inputs: total cost' column shows each household's total cost ('raw' [i.e. not standardised for household size], and in Ethiopian birr) of purchasing agricultural inputs during the study period. The 'Livestock inputs: total cost' column shows each household's total cost ('raw' [i.e. not standardised for household size], and in Ethiopian birr) of purchasing livestock inputs during the study period. 6. 'ART DI_AEs, HHs included in final analysis' There are many potential types of output from the open-IHM data; this folder contains what is probably the most crucial type - disposable income per adult equivalent (DI/AE) - for the households included in the final Assesing Rural Transformations (ART) analysis. The DI/AEs here have all been produced from data equivalent to that in the anonymised open-IHM databases, following the relevant steps in the 'Instructions for accessing and interpreting IHM data from the ART project.docx' document (including changes to the food energy values and computer date where necessary, as well as factoring in appropriate non-automated input costs), and without any adjustment for inflation. By carefully following the instructions above and in the 'Instructions for accessing and interpreting IHM data from the ART project.docx' document carefully, you should therefore be able to produce the same DI/AEs for these households from their open-IHM datasets, as well as DI/AEs for the households included in these datasets but not included in the final ART analysis. However, because all endline cash data used in the final analysis was adjusted for compound inflation, the endline DI/AEs here are different to the follow-up DI/AEs analysed. The compound inflation rates used are provided below and in the working papers. Note that all references to 'DI_AE' are intended to be read as 'DI/AE' - the reason for using the underscore here is the inability to use a forward-slash in the name of a file or folder. 6a. 'Karonga district 2013 DI_AE.xls' This spreadsheet contains the baseline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Karonga district, Malawi. The data for this is in the 'Maulunga 2013 Anonymised.ihm' and 'Mkango 2013 Anonymised.ihm' open-IHM databases. To replicate the output here from these databases, you would also need to add 80,000 to the household ID numbers/codes of the households from Maulunga, and to add 90,000 to the household ID numbers/codes of the households from Mkango. In the spreadsheet: The 'New HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's baseline annual disposable income per adult equivalent (i.e. standardised for household size, and in Malawian kwacha) over the course of the study period. These disposable incomes are not adjusted for inflation; as baseline figures, they were not adjusted for inflation in the final analysis either. 6b. 'Karonga district 2015 DI_AE.xls' This spreadsheet contains the endline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Karonga district, Malawi. The data for this is in the 'Maulunga 2015 Anonymised.ihm' and 'Mkango 2013 Anonymised.ihm' open-IHM databases. To replicate the output here from these databases, you would also need to add 80,000 to the household ID numbers/codes of the households from Maulunga, and to add 90,000 to the household ID numbers/codes of the households from Mkango. In the spreadsheet: The 'New HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's endline annual disposable income per adult equivalent (i.e. standardised for household size, and in Malawian kwacha) over the course of the study period. These disposable incomes are not adjusted for inflation; in the final analysis they were adjusted for compound inflation of 47.59836% (lowering positive disposable incomes and raising negative disposable incomes). 6c. 'Lilongwe district 2012 DI_AE.xls' This spreadsheet contains the baseline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Lilongwe district, Malawi. The data for this is in the 'Masumbankhunda 2012 anonymised.ihm' open-IHM database. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's baseline annual disposable income per adult equivalent (i.e. standardised for household size, and in Malawian kwacha) over the course of the study period. These disposable incomes are not adjusted for inflation; as baseline figures, they were not adjusted for inflation in the final analysis either. 6d. 'Lilongwe district 2014 DI_AE.xls' This spreadsheet contains the endline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Lilongwe district, Malawi. The data for this is in the 'Masumbankhunda 2014 Anonymised.ihm' open-IHM database. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's endline annual disposable income per adult equivalent (i.e. standardised for household size, and in Malawian kwacha) over the course of the study period. These disposable incomes are not adjusted for inflation; in the final analysis they were adjusted for compound inflation of 51.00442% (lowering positive disposable incomes and raising negative disposable incomes). 6e. 'Oromia 2012 DI_AE, minus input costs.xls' This spreadsheet contains the baseline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Oromia, Ethiopia. The initial data for this is in the 'Golla Village Anonymised.ihm' and 'Malt Barley Beneficiaries Anonymised.ihm' open-IHM databases. To replicate the output here from these databases, you would also need to use the input costs from the 'Assela (Oromia) 2012 'Golla village' non-automated input costs.xlsx' and the 'Assela (Oromia) 2012 'malt barley beneficiaries' non-automated input costs.xlsx' spreadsheets - first dividing these input costs by the appropriate households' numbers of adult equivalents (see step 3.2 in the 'Instructions for accessing and interpreting IHM data from the ART project.docx' document) and then subtracting the results from the appropriate DI/AEs. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's baseline annual disposable income per adult equivalent (i.e. standardised for household size, and in Ethiopian birr) over the course of the study period. These disposable incomes are not adjusted for inflation; as baseline figures, they were not adjusted for inflation in the final analysis either. 6f. 'Oromia 2014 DI_AE, minus input costs.xls' This spreadsheet contains the endline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Oromia, Ethiopia. The initial data for this is in the 'malt barley assela round 3 anonymised.ihm' open-IHM database. To replicate the output here from this database, you would also need to use the input costs from the 'Assela (Oromia) 2014 non-automated input costs.xlsx' spreadsheet - first dividing these input costs by the appropriate households' numbers of adult equivalents (see step 3.2 in the 'Instructions for accessing and interpreting IHM data from the ART project.docx' document) and then subtracting the results from the appropriate DI/AEs. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's endline annual disposable income per adult equivalent (i.e. standardised for household size, and in Ethiopian birr) over the course of the study period. These disposable incomes are not adjusted for inflation; in the final analysis they were adjusted for compound inflation of 21.5486712926119% (lowering positive disposable incomes and raising negative disposable incomes). 6g. 'Tigray 2014 DI_AE, minus input costs.xls' This spreadsheet contains the baseline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Tigray, Ethiopia. The initial data for this is in the 'Tigray 2014 Anonymised.ihm' open-IHM database. To replicate the output here from this database, you would also need to use the input costs from the 'Tigray 2014 non-automated input costs.xlsx' spreadsheet - first dividing these input costs by the appropriate households' numbers of adult equivalents (see step 3.2 in the 'Instructions for accessing and interpreting IHM data from the ART project.docx' document) and then subtracting the results from the appropriate DI/AEs. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's baseline annual disposable income per adult equivalent (i.e. standardised for household size, and in Ethiopian birr) over the course of the study period. These disposable incomes are not adjusted for inflation; as baseline figures, they were not adjusted for inflation in the final analysis either. 6h. 'Tigray 2015 DI_AE, minus auto input costs.xls' This spreadsheet contains the endline disposable incomes per adult equivalent (DI/AEs) for the households included in the final ART IHM analysis for Tigray, Ethiopia. The data for this is in the 'Tigray 2015 Anonymised.ihm' open-IHM database, which is set up with input costs automatically subtracted from relevant output. In the spreadsheet: The 'HH no.' column shows the final IHM identifying household number/code of each household. The 'DI/AE' column shows each household's endline annual disposable income per adult equivalent (i.e. standardised for household size, and in Ethiopian birr) over the course of the study period. These disposable incomes are not adjusted for inflation; in the final analysis they were adjusted for year-on-year inflation of 3.33538474414852% (lowering positive disposable incomes and raising negative disposable incomes).