3.6. Reliability and validity
According to Schwab et al. (2014), validity and
reliability in research design refer to the need to ensure that concepts used
in the study measure what they are intended to and that this measurement is
consistent and stable for all respondents.
The researcher used professionals or experts in the strategy
field to assess the concept the instrument is trying to measure and also
determine whether the set of items or checklist accurately represents the
concepts under the study.
For reliability, Cronbach's Alpha is a reliability coefficient
that indicates how well items in a set are positively correlated to one
another. Generally, reliability of 0.7 to 1.0 is considered acceptable. For
this study, an alpha coefficient is 0.81 which is greater than 0.7, i.e.
considered reliable.
3.7 Data processing and
analysis
Data processing is concerned with classifying responses into
meaningful categories called codes (Roth, 1989: 58). It consists of editing,
schedules and coding the responds. The data processing began with editing,
coding and finally ended with tabulation.
Nachmias (1976: 143) assert that, data processing is a link
between data collection and data analysis. It involves the transformation of
the observation gathered from the fields into the system of categories and the
transformation of these categories into codes and amenable to quantitative
analysis and tabulation. Not all data can be presented in their entity.
The variables to be presented are those most central to the
goals of the study; generally, variables include specifically in the research
questions (Bailey, 1978: 321).
Data collected were really in arrow form. It was not easy to
the researcher to present the findings of the study. Therefore, the researcher
had technically processed data before proper analysis was done so as to become
more meaningful for interpretation. Data processing was done in accordance with
general and specific objectives of the research study.
After carrying out interviews with respondents and
questionnaire were collected, they were edited and the information was arranged
in a meaningful and organized form by coding it. The data were analyzed by
using statistical package for social science (SPSS).
3.7.1 Editing
According to Daniel and Gates (1991: 387), editing is the
process of going through the questionnaire to ensure that the `skip patterns'
were followed and required questions are filled out. Editing involves the
inspection and if necessary, connections of each questionnaire or observation
form; the basic purpose of editing is to impose some minimum quality standards
on the raw data (Churchill, 1992:608).
3.7.2 Coding
According to Churchill (1992: 612), coding is the procedure by
which data are categorized. Through coding, the raw data are transformed into
symbols usually numerals that may be tabulated and counted. The transformation
is not automatic; however, it involves judgment on the part of coder. Moses and
Kalton (1971: 415), states that the purpose of coding in the survey is to
classify the answers acquired were coded and tallies used to determine the
frequencies of each response. Similar responses would be grouped according to
their different categories.
This helped the researcher to know for instance the percentage
of responses on whether there is positive effect of e-procurement
implementation in Rwandan Hospitals especially in Kabaya District Hosptal.
| 
