Validity
It indicates the extent to which an instrument measures what
it is supposed to measure. Six experts in the field have checked the
questionnaire for the consistency of the items, conciseness, intelligibility
and clarity. The checking of items, consistence, relevance, clarity and
ambiguity; pretesting was done in two health centers that were not part of the
target population.
Their input helped to ensure that the
instrument measured adequately what it is intended to measure. The researcher
used CVR (Content Validity Ratio) where the expert will agree with the items.
The formula to be used is: CVR = (E -N/2) / (N/2)
Where E: number of who rated the object or
person in question; N: total number of expert. CVR can measure
between -1.0 and 1.0. The closer to 1.0 the CVR is, the more essential the
object is considered to be. Conversely, the closer to -1.0 the CVR is, the more
non-essential it is.
The research instrument was valid when the CVR is 0.6 or above
indicated the extent to which an instrument measures what it is supposed to
measure. A supervisor was always consulted for checking the items, consistence,
relevance and clarity.
Reliability
Twenty CHWs from Kabarondo Health center were randomly
selected for testing research instrument, the estimation of reliability will be
ascertained by a pilot testing of the instrument and applying Cronbach's Alpha
coefficient by means of a Statistical Package for Social Sciences (Gal,et
al,2009). Cronbach's Alpha coefficient will be used to measure internal
consistency of the research tool. The instrument are reliable when the results
of twenty respondents give an alpha coefficient of > 0.7 (Gal, et al,
2009).
Data Collection
Procedure
The researcher obtained a letter of introduction from Bugema
University, Graduate school to the Director of Rwinkwavu District Hospital. The
researcher submitted the letter in person to the office of Rwinkwavu District
Hospital Director and upon authorization; the researcher made an appointment
through the community health workers in charge of health center level to
confirm when he could visit to collect data as community health workers
involved in this study live in different areas. The questionnaire was given to
the respondents after ensuring them that the information given will be kept
confidentially and would be used only for academic and research purpose.
The researcher ensured voluntary participation of respondents
to be clearly informed about the objective and benefits of the study, the
confidentiality of records was protected and no name of respondents were asked
during the data collection.
Data Analysis
After a successful data collection exercise, the researcher
coded and entered data, tabulated and interpreted the findings. For
quantitative data, the computer package, SPSS was used to analyze and interpret
the data. Descriptive statistic including frequency and percentage was used to
answer objectives one, the mean and standard deviation was used to answer the
objective two and three. Linear regression logistic was used to analyze the
objectives four that was to establish the influence of incentives on improving
maternal and newborn health services. Descriptive statistics allowed the
researcher to reduce bias and estimate sampling errors and precision of the
estimates derived through statistical calculation. Data collected from the
document analysis was analyzed manually and results were used to supplement and
support the findings from the main instrument.
| 
