A set of statements will be developed to encompass the housing affordability index. These statements will be considered to be a complete set for measuring housing affordability the dependent variable. The study will adopt a five point scale ranging from 1 to 5 with a score of 1=strongly disagree, 2=disagree ,3=neutral 4=agree and 5=strongly agree. Mutisya (2015) and Ngigi (2016) used a five point likert scale to measure housing affordability in Kenya, this study also will adopt five point likert scale. For this case, Ten (10) statements will be used to measure housing affordability index (question 52 .appendix.3.) Further Guideline index will be measured using Eight (8) statements (question53 .appendix3) Housing poverty Index will be measured using Five(5) statements (question 54 appendix3) self-sufficiency measure Index will be measured using Five(5) statements (question 55.appendix3) Amenity-Based Housing index will be measured using Five (5) statements (question 56 .appendix)
These housing affordability measures will be evaluated by one criterion on three dimensions. The only criterion to evaluate them is whether the selected measurement can accurately describe the relationship between housing and those who live in it. The three dimensions of the criterion are considered to have the same weight. These three dimensions are standards, accuracy of information, and comprehensiveness.
The study will adopt a seven point scale ranging from 1 to 5 point scale 1=strongly disagree, 2=disagree ,3=neutral 4=agree and 5=strongly agree. as in previous cases. As is the case of other indices computed and weighted in the same way as follows: Three (3) statements of standards will be developed to measure housing affordability (question 57 appendix 3) Three (3) statements of accuracy of information will be developed to measure housing affordability (question 58.appendix 3) Three (3) statements of comprehensiveness will be developed to measure housing affordability (question59 .appendix 3) and subjected also to factor analysis for data reduction. For each respondent, the sum of weights on respective statements will be added up and divided by the number of statements in that particular subset that remained after data reduction.
3.6.4 Specification of the model
The pro bit model will be used to establish whether housing co-operative deliver affordable housing or not. Affordable housing will be dependent variable. Affordability of housing will be observed as 1 if household had access to affordable housing and 0, otherwise. The model will be as follows.