Eastern Zone DCSA

Recommendations

Info

1.1. Overview As part of its continual assessment of issues directly affecting community security and safety, OCVP conducted extensive primary data collection in several districts of Mogadishu: Karaan, Yakshiid, Heliwaa and Shibis districts which comprise the EASTERN ZONE. In order to gain a comprehensive understanding of the thematic areas under investigation a mixed method approach was employed in order to allow the research team to triangulate information uncovered in both the data collection phase and subsequent analysis. The household surveys aimed to obtain a representative picture of the thematic areas under study and Key Informant Interviews were used to probe deeper into, and cross-validate, issues that emerged from the Focus Group Discussions. 1. METHODOLOGY enabled the application of a sampling formula 2 to determine a representative sample size for each district. Taking into account certain statistical parameters such as the level of confidence desired (95%), sample design effect (1.5), margin of error (+ or – 5%) and the assumption that some security correlation of (0.3) existed within the sub-divisions. The p-value of 0.3 in the formula assumes security correlation above a random normal distribution of 0.5 within the district clusters. This is a reasonable assumption based on the topic of the survey – respondents within the districts of Mogadishu are likely to possess a correlating relationship between their perceptions based of the surrounding security environment. This is further backed on OCVP’s past experience in which those from the same area tend to exhibit similar perceptions on security, justice, governance and conflict and violence. The quantitative data was analysed by the OCVP Research and Analysis team using Statistical Package for Social Science (SPSS) version 21, after the data had been collated and cleaned by Swisspeaks - an independent quality control firm. The qualitative data was subjected to thematic analysis, using a largely inductive process. The main themes of coding were developed before the mission in line with the hypothesis but further coding was inductively included during analysis. Finally a total of 483 household questionnaires were calculated for the zone (although not all responses were collected, see qualitative limitations at section 1.6), and then a calculated ratio (number of households in the district over the total number of households in the zone multiplied by the calculated sample size for the zone) was employed to come up with the number of households that should represent each district. 1.2. Sampling Methodology A district household survey provided by Heritage Institute for Policy Studies (HIPS) and later cross-checked with the regional authorities 2 See Annex 5.1 1
1.3. Household Survey A face-to-face quantitative survey was conducted in which questions relating to personal demographics, security, justice and governance provision, and conflict and violence were asked. OCVP researchers trained eight data enumerators (4 male and 4 female) on the questionnaire methodology and the use of smart-phones for data collection from February 24 th to 25 th 2014. During the training the enumerators and OCVP researchers discussed the use of smartphones for data collection and the potential risks involved and finally agreed on measures to mitigate the identified risks (including an explanation on the use of the device and the request for the respondent’s prior consent). The <strong>Eastern</strong> <strong>Zone</strong> is comprised of the eastern districts of Mogadishu and was observed to be relatively peaceful at the time of the data collection - except Heliwaa district which was deemed to be very insecure. Based on previous experience and data collection exercises in the other zones of Mogadishu, researchers were aware of the possible existence of ‘hot spots’ – areas where the security situation was particularly volatile and data collection was deemed to be higher risk. Adhering to the Systematic Random principle (SRM) employed in the central zone would have been impractical and potentially dangerous; therefore researchers were obliged to recruit local facilitators from the district to avoid hot spots and potentially violent elements of the community. These facilitators possessed a profound knowledge of the district and a pre-existing relationship with the local population. The local facilitators advised on the enumerators’ movements and personal security. In addition to that, they facilitate the collection of some household questionnaires based on their personal connections and acquaintances. This has resulted, however, in the random systematic selection of the households being compromised and somewhat undermined the validity of the data collected. The eight trained enumerators supported by the two local facilitators (recruited from the district administrations) under the supervision of two OCVP researchers were able to identify the divisions, subdivisions and neighbourhoods in each of the districts and then agreed on a starting point for each of the enumerators. The local facilitators were also able to identify boundaries between the subdivisions and neighbourhoods. Where appropriate, enumerators randomly selected any 5th household and interviewed one respondent above the age of 18 or would interview the household respondents based on the local facilitator’s identification and/or advice. A gender balance was emphasised across the entire survey. The data collection was carried out until the pre-calculated number of households in the district were covered (though 20 out 304 households in Yakshiid and the whole of Heliwaa district - 83 households - were not surveyed). Compromising the sampling probability method was seen as a necessary trade-off for successful data collection and the safety of the enumerators. The 8 enumerators collected 380 household observations in three districts: Karaan, Yakshiid and Shibis from 19th to 23rd March 2014. The data was collected in Somali language with the use of smartphones which was subsequently uploaded onto the SwissPeaks database when an internet connection was available. 2
Page 2 and 3: Mogadishu Conflict and Security Ass
Page 4 and 5: ACKNOWLEDGMENT The Observatory of C
Page 6 and 7: Contents ACKNOWLEDGMENT ...........
Page 8 and 9: Table of Figures Figure 1: Number o
Page 10 and 11: The Study Area Map Map of Mogadishu
Page 12 and 13: police was very high. The qualitati
Page 16 and 17: Figure 1: Number of male and female
Page 18 and 19: household observations were lost. H
Page 20 and 21: Most of the respondents (74%) (n=28
Page 22 and 23: Figure 8: Respondents’ awareness
Page 24 and 25: Figure 11: Respondents’ account o
Page 26 and 27: Figure 15: Respondents’ perceptio
Page 28 and 29: The Karaan District Deputy Commissi
Page 30 and 31: stations, outposts, police officers
Page 32 and 33: A proportion of respondents reporte
Page 34 and 35: Figure 27: Respondents’ estimatio
Page 36 and 37: there are also buildings with five
Page 38 and 39: Figure 36: Respondents’ account o
Page 40 and 41: Figure 39: Respondents’ trust of
Page 42 and 43: Figure 43: Respondents' perception
Page 44 and 45: 3.4.2. Performance of the Governanc
Page 46 and 47: Figure 49: Respondents’ perceptio
Page 48 and 49: 4. CONCLUSION Though there has been
Page 50 and 51: Annex 5.2: GLOSSARY OF TERMS Access

Eastern Zone DCSA

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?