Kilic, T., and Moylan, H. (2016). “Methodological experiment on measuring asset ownership from a gender perspective (MEXA): technical report.” Washington, DC: World Bank

Selected Land policy normative documents

Africa Union, African Development bank and United Nations Economic Commission for Africa (1999). Land Policy in Africa: A Framework to Strengthen Land Rights, Enhance Productivity and Secure Livelihoods. Available at: https://www.uneca.org/publications/framework-and-guidelines-landpolicy-africa

Africa Union, African Development bank and United Nations Economic Commission for Africa (2014). Guiding Principles on Large-Scale Land-Based Investment in Africa. Nairobi. Available at: https://www.uneca.org/sites/default/files/PublicationFiles/guiding_principles_eng_rev_era_size.pdf

Food and Agriculture Organization of the United Nations (2012). Voluntary Guidelines on the Responsible Governance of Tenure of Land, Fisheries and Forests in the Context of National Food Security. Available at: http://www.fao.org/docrep/016/i2801e/i2801e.pdf

Proceedings EGMs for SDG 1.4.2

Expert Group Meetings on methodology development using survey data: https://gltn.net/home/download/international-expert-group-meeting-on-land-tenure-security-to-develop-a-set-of-household-survey-questions-for-monitoring-sdg-indicator-1-4-2/?wpdmdl=111

Expert Group Meetings on methodology development using administrative data (http://documents.worldbank.org/curated/en/482991505367111149/pdf/119691-WP-P095390-PUBLIC-SDGEGMproceedingsuseofadministrativedatalandagencies.pdf)

Consolidated essential questions land module for 1.4.2 and 5.a.1 (FAO, UN-Habitat, UN Women, World Bank). Module for individual interviewing under preparation; Version for household surveys with proxy respondents; available at: http://documents.worldbank.org/curated/en/812621505371556739/Land-tenure-module-essential-questions-for-data-collection-for-1-4-2-and-5-a-1).

ANNEX:

Full methodology development narrative (including list of pilot countries, data and other results from pilot studies)

METHODOLOGICAL DEVELOPMENT

Global consultations on the methodological developments of this indicator were conducted with a diverse range of participants and partners. The custodian agencies, working directly with NSOs and land agencies, developed tools and capacity development packs, followed by computation of data points for relevant variables for this indicator for several countries on a pilot basis, using existing data sources from nationally representative surveys and census and, in exceptional cases, rigorous impact evaluations without national coverage.

Methodology development and piloting results

Formulas and combining different elements

The process used for methodology development are presented above. As discussed in detail there, indicator 1.4.2 comprises two parts: (A) measures the incidence of adults with legally recognized documentation over land among the total adult population; while (B) focuses on the incidence of adults who report having perceived secure rights to land among the adult population. Part (A) and part (B) provide two complementary data sets on security of tenure rights, needed for measuring the indicator.

Part (A):

$\frac{P e o p l e (A d u l t) w i t h l e g a l l y r e c o g n i z e d d o c u m e n t a t i o n o v e r l a n d}{T o t a l a d u l t p o p u l a t i o n }$ X 100

Part (B):

$\frac{P e o p l e (a d u l t) w h o p e r c e i v e t h e i r r i g h t s a s s e c u r e}{T o t a l a d u l t p o p u l a t i o n }$ X 100

The computation formula has built in system for computing the individual components of this indicator.

Where survey data are collected separately for agricultural and residential land, double counting is avoided by adjusting for households that access both types of land simultaneously.
Strata title: cases where a residence is in an apartment building, the rights to the residency are counted as rights to the land.
For purposes of retrospective data collection, parcels that are already affected by a dispute are also included in the reporting below on fear for involuntary loss of land.

As required by the indicator definition, any component can be disaggregated by gender and tenure type.

The national censuses or household surveys by the national statistical system were used to assess the number of people to access any land either through individual or joint ownership or via rental. Gender was calculated from surveys or calculated by land agencies using administrative data.

More detailed technical issues, e.g. ways to deal with proxy reporting by one member of the household on and when and how administrative data can be used are explained in the draft meta data for SDG 1.4.2.

Piloting results

Results from applying the methodology to select data are summarized in Table 1. They demonstrate not only the viability of the methodology, including the scope for how survey and administrative data to complement each other in a useful way, as well as the ability to derive a meaningful and actionable indicator. Rather than discussing substantive implications and actionability at country level, we focus on cross-cutting and data issues, illustrating in particular how different data sources can usefully complement each other.

Table 1: Selected countries with data on indicator 1.4.2

Country/Region	Data Source(s)	Year	Land access via		Formal	Perceived	Index	Gender
			Ownership	Rental	Document	Security
Africa
Benin	INSAE, MCC & Admin	2011	0.809	0.047	0.113	0.903	0.51	0.123
Lesotho	MCC	2013	0.914	0.029	0.611	0.929	0.77
Mozambique	INE	2011	0.882	0.033	0.498	0.811	0.65	0.112
Malawi	NBS	2015	0.868	0.023	0.019	0.697	0.36	0.226
Nigeria	NBS	2013	0.741	0.025	0.021	0.741	0.38	0.162
Rwanda	LSMS-ISA & admin data	2015	0.886	0.002	0.858	0.969	0.91	0.864
Tanzania	LSMS-ISA	2013	0.839	0.123	0.250	0.960	0.61	0.339
Uganda	LSMS-ISA	2014	0.902	0.080	0.080	0.919	0.50	0.525
Asia
Korea, Rep.	Census & Admin	2016	0.723	0.237	1.000	0.960	0.98
Mongolia	MCC-SHPS	2012	0.809	0.163	0.654	0.966	0.81	0.268
Americas
Costa Rica	Census & Admin	2011	0.699	0.279	1.000	0.978	0.99
Europe
Belgium	Census & Admin	2011	0.628	0.362	0.948	0.939	0.94	0.543
Netherlands	Census & Admin	2011	0.539	0.429	1.000	0.968	0.98	0.640
Oceania
New Zealand	Census & Admin	2013	0.607	0.327	0.990	0.925	0.96

Source: SDG Indicator 1.4.2 Global Database

Selected data comments

The data on formal documentation of land rights of the indicator are self-reported from household surveys for most low-income countries or from land agencies’ records. Ranging from less about 2% in Malawi and Nigeria to full coverage in Costa Rica, the Netherlands, and Korea, there is enormous variation in this part of the indicator across countries.

Data on perceived tenure security, which is from survey data and self-reported for all except the European countries and Costa Rica. In the latter case, we used the share of the population who, according to the population census or household surveys, either report owning or renting land or their residence to represent the share of the population who enjoy legally recognized documentation and by implication whose tenure is legally secure.

Benin: Administrative data indicate that the population have received individual documents based on the Plan Foncier Rural, but also suggest that, with about 12% of documents registered in their name, women have not benefited to the extent that may be expected.

Costa Rica: Administrative records suggest that in Costa Rica, all land is covered with land records. But census data indicate that some 2.5% of the population still suffers from precarious tenure, highlighting that even in cases where administrative data are available, they need to be linked to population-based evidence to give a fuller picture.

Malawi: Although the Government is engaged in an ambitious effort to digitize available records that would provide a basis for better land administration and reporting, only information from household survey data is available. The survey data point to high levels of tenure insecurity that are mostly gender related.

Nigeria: As a federal country, several states in Nigeria have administrative data that are of sufficient quality for reporting for the pilot. Data suggest that insecurity is high and, in many cases, is caused by the state due to expropriation.

Rwanda: A representative survey is available and points towards high levels of tenure security. However, information on the gender distribution of legally recognized documentation can be more reliably obtained from administrative data that show that more than 86% of women have land registered in their name either individually or jointly and perception of tenure security is high.

The Netherlands: This case shows that administrative data can be gender-disaggregated, and that in an advanced economy the share of individuals accessing land through various forms of institutions is high.

The Questionnaire Module with essential question for reporting on SDG 1.4.2

The module with essential questions for reporting on 1.4.2 is discussed in detail below. The module is developed by the UN Habitat and the World Bank, together with FAO and UN Women, with inputs from other stakeholders through GDWGL and GLII, and supported by the Living Standard Measurement Survey (LSMS) team.

The results of the EDGE project and other recent evidence suggest that individual level data collection is preferred to potential proxy respondents (where feasible).

Because of the scalability benefits of collecting data for both indicators simultaneously, the module is designed to provide the data required to compute indicators 1.4.2 as well as 5.a.1. Only the essential questions for indicator computation are included, however, the module may be expanded upon as needed by NSOs to address a wider range of land tenure issues relevant at the country level.

The module example, appended to this note is designed as a household level questionnaire in which a full roster of parcels is collected at the household level and the module is then implemented for each parcel, where the respondent is the most knowledgeable household member for the given parcel.

The module incorporates lessons learnt from methodological experiments1, as well as from implementation at national scale by the national Statistical Office of Malawi in its 2016/17 Integrated Household Survey (IHS4). The IHS4 interviewed 12,480 cross-sectional households across 780 EAs, and in parallel, revisited a national sub-sample of 2,516 households that had been previously interviewed in 2010 and 2013. As part of the IHS4 panel component, the survey administered up to 4 adult individual interviews per household. The modules asked separately questions regarding (i) reported ownership, (ii) economic ownership, (iii) documented ownership, and rights to (iv) sell, (v) bequeath, (vi) use as collateral, (vii) rent out, and (viii) make improvements/ invest.

Indicator 1.4.2 considers two aspects of tenure security: documentation and perception. Only documentation that is official, and therefore provides legally protected tenure rights, is considered under indicator 1.4.2. That is combined with perception of tenure security, which is captured through the respondent-estimated probability of involuntary loss of land rights in the next five-year period and the reported right to bequeath.

While the module has been carefully designed to be as universal as possible to maintain comparability of the computed indicator across time and space, certain questions, marked in the questionnaire, will require customization at the country level. Customization cannot be avoided in full due to the varying legal systems and land tenure arrangements across countries. Collection of metadata, including the identification of legally recognized documentation in the particular country context must take place prior to implementation of the module.

The Questionnaire Module

The questionnaire module assumes a survey that has households as the unit of enumeration and analysis, and where a household roster is used to identify household members and collected basic information on their demographics including age and gender. In this process, each household member is assigned a unique identifier (HHID). In the Annex, the questions are color-coded to identify those required for indicator 1.4.2 only, for indicator 5.a.1 only, for both indicators, and those included for disaggregation or other analytical purposes. In what follows, practical issues for implementation are discussed, and explanatory notes on each individual question presented.

Scalability & Up-Take

While it is to be expected that the module will be usually implemented in conjunction with a larger survey operation, nothing prevents users to implement it independently. Implementing the module in the context of multi-topic surveys will increase its analytical value as, beyond generating an indicator for the SDG monitoring process, countries would be in a position to explore how land tenure issues relate to other development outcomes, including other SDG goals. The custodians foresee implementing the module as part of Living Standard Measurement Study (LSMS) surveys and the Urban Inequality Survey (UIS), and will be discussed with the USAID-funded Demographic and Health Surveys (DHS) programme, and UNICEF Multiple Indicator Cluster Surveys (MICS). Any nationally representative sample survey can of course become a vehicle for implementing the module. The custodians envisage working with National Statistical Offices to engage in dissemination and capacity development, as integrating the module in national statistical programs is the only viable way to ensure sustainability of the data collection process and ownership of the results by countries.

Implementation Method

The questionnaire module has been designed for paper assisted personal interviewing (PAPI) implementation to have the widest reach. However, an electronic version of the questionnaire will be created by the custodians for use in computer assisted personal interviewing. The application will be created using the World Bank’s open access CAPI platform, Survey Solutions (solutions.worldbank. org), and will be made publicly available. The CAPI application can be customized from the base module as necessary. Implementation of the module via CAPI is recommended, as this can minimize data entry errors, allow for more immediate data review and analysis, and enable quick use of photo aids (which can improve data quality).

Before Going to the Field: Collecting Metadata

In this context, metadata refers to the classification of land documentation into legally recognized and unrecognized types as defined for indicator 1.4.2. The metadata will vary by country and will therefore, need to be released along with the computation of the indicator for transparency, and update in the case of changes in the regulatory frameworks. The metadata will identify which types of documentation are legally recognized, and therefore, what constitutes secure tenure. Questions on unrecognized and/or informal documentation can be asked separately, but is not considered in the computation of Indicator 1.4.2.

Question-by-Question Guidance

The implementation of the questionnaire included in the Annex is fairly intuitive, yet it is recommended that prior to its implementation, adequate training is provided and an enumerator manual is produced to guide data collection, including with images of the range of tenure related documentation in use by land holders. Detailed explanatory notes on each question are found below, which can be used to develop such manuals. Where customization is necessary, this is indicated. Annex I also indicates skip patterns (indicated by the arrow sign ‘>>’).

Guidance for sample questionnaire annex 1 (household survey with parcel roster)

RESPONDENT ID:	The respondent ID is the ID of the person responding for the respective parcel, recorded from the household roster. The respondent should be the most knowledgeable household member for each parcel. Therefore, the respondent may differ for each parcel. The optimal respondent should be identified through a discussion amongst the enumerator and all adult members of the household (or as many as possible) prior to beginning the module. During this meeting, the full roster of parcels should be recorded and the optimal respondent identified for each.
Q1:	The roster of parcels should contain all parcels used by, owned by, or occupied by any household member(s) at the time of the interview. Alternatively, a single set date could be identified for a given survey. This option is especially applicable in when fieldwork is conducted over an extended period of time (such as a 12-month rolling fieldwork design). The first parcel listed should be the parcel on which the household resides. The parcel name must be unique to each parcel, as it will be used to refer to the specific parcel throughout the remainder of the module. In the case of panel surveys, or surveys with multiple visits, parcel names referring to a crop grown, for example, should be avoided as that may change over time.
Q2:	Parcel area has been included in the module to allow for disaggregation of the indicator (for example, for smallholder farmers only). Farmer estimation of parcel area should be collected for all parcels. Additionally, GPS measurement of parcels is strongly advised, wherever feasible. Recent evidence points to systematic bias in farmer estimates of land area1. Land area units must be customized for the country context.
Q3:	Parcel acquisition type is used as a filter question for the following questions, allowing for maximum efficiency in skipping questions where possible. Response code to be reviewed in light of the country context.
Q4:	The tenure system of the parcel is used to disaggregate indicator 1.4.2. Response codes to be reviewed in light of the country context.
Q5:	The primary use of the current parcel is used to disaggregate indicator 1.4.2, and to identify land subject to indicator 5.a.1, which pertains to agricultural land. In some cases, such as when land is rented out, the actual use may not be known, hence the inclusion of the “Don’t Know” response. However, wherever possible, the actual use of the land, rather than current ownership or use arrangements, should be recorded.
Q6:	Question 6 identifies the owner(s) or use right holder(s) of the parcel, as reported by the respondent. Multiple household members may be listed, as joint ownership/use right holding is common.
Q7:	This module only seeks to identify the possession of documents that are pre-determined to be legally recognized in the given context. Question 7, therefore, asks about the possession of documents from a specific government agency(ies). Examples of relevant documents are embedded in the question to provide context to the respondent and to clarify that documents other than title deeds are relevant. The government agency(ies) and example documents embedded in the question must be customized for the country context. Refer to the section above on metadata for guidance on determining what is to be classified as legally recognized.
Q8:	If the response to Question 7 is “yes”, question 8 is answered to record the specific type of documents held by the household, and which members are named on each. Codes must be customized at country level to include all legally recognized documents (as determined through the pre-survey preparation of metadata). Rental contracts of some form should be included, as long as rights are legally protected. To minimize errors in naming and classifying documents, a photo aid containing an image of all legally recognized documents should be constructed and shown to the respondent. The integration of visual aids (e.g. a photo of an actual document of the reproduction of a facsimile) is most easily done in a CAPI application, but can also be integrated in traditional PAPI interviews.
Q9:	The right to sell the parcel is captured in questions 9 and 10. Question 9 is a filter question, asking if any household member has the right to sell the parcel, either alone or jointly. That is, if any household member has the right to sell (or believes they have the right to sell) whether that be alone or with the approval/signature/etc. of another person either within or outside the household, the respond should be “yes”. This question is skipped for parcels acquired through short-term rentals (<3 years) and sharecropping-in. Questions on the right to sell are used for computation of indicator 5.a.1 only.
Q10:	List the ID codes of the household members that have the right to sell the parcel. If there are any external members that have the right to sell, enter the code accordingly. This question is skipped for parcels acquired through short-term rentals (<3 years) and sharecropping-in.
Q11:	The right to bequeath the parcel is captured in questions 10 and 11. Question 10 is a filter question, asking if any household member has the right to bequeath the parcel, either alone or jointly. That is, if any household member has the right to bequeath (or believes they have the right to bequeath) whether that be alone or with the approval/signature/etc. of another person either within or outside the household, the respond should be “yes”. This question is skipped for parcels acquired through short-term rentals (<3 years) and sharecropping-in. Here, bequeath is defined as the ability to transfer rights to the parcel either in life or in death.
Q12:	List the ID codes of the household members that have the right to bequeath the parcel. If there are any external members that have the right to bequeath, enter the code accordingly. This question is skipped for parcels acquired through short-term rentals (<3 years) and sharecropping-in.
Q13:	Question 13 identifies the likelihood of involuntarily losing ownership/use rights to the parcel in the next five years. Responses are made on a scale from 1 to 7, with 1 being not at all likely and 7 being extremely likely. This question is asked about each owner/use right holder separately that was identified in Question 6 (but asked all to the same parcel-level respondent). This formulation of the question allows for the observance of intra-household insecurity, for example involuntary transfer of rights from female to male household members. For parcels acquired through short-term rental (<3 years), the question will be asked for likelihood of involuntary loss in the remaining duration of the contract.

G. Carletto, S. Gourlay, S. Murray, and A. Zezza (2016), Land Area Measurement in Household Surveys, Washington DC, The World Bank.

Hesin internet-kagin er ikki longur stuðlaður

Rætta metadátur: 1.4.2