Methodology is a cornerstone of research as it looks at how the researcher constructs knowledge in
his/her works that is, how the research is conducted. Within the broader research methodology framework,
research is mostly conducted to collect data on any particular phenomenon and critical to gathering data is the
sample selection. Therefore sampling becomes the most critical part of any population survey based research.
This is because when conducting ‘research study’, in most surveys, access to the entire population is impossible,
therefore applying the right sample using appropriate sampling technique gives confident results that reflect
extremely closely to those that would have been obtained had the entire population provided the data. Given that
selecting a sample is a crucial step in any research project as is rarely practical, efficient or ethical to study the
whole population as aforementioned. This paper looks at the various literatures and techniques applied,
arguments raised for preferring a particular approach as opposed to other and how various researches have
approached the issue of what constitutes an appropriate sample size. The literature reveals different divergent
views on sampling selection and sample size and this position has prompted this paper to investigate further the
sampling selection approaches and ultimately develop and present a new population targeted survey based
sampling model (NPTSBS- Model), which model is based on the methods, approaches and techniques employed
by the researcher when carrying out data collection for a PhD programme on eGovernment implementation for
Botswana government. The NPTSBS model is a hybrid of probability and non probability models; it comprises
cluster, stratified random and purposive sampling techniques. The fundamental question being addressed in the
model is that given the various characteristics of a population, “How can the research ensure a balanced
representative or most inclusion of the various composite characteristics of the population within the envisaged
sample selection.’’ Although the new model is developed based on an eGovernment research survey, it is
however broadly flexible, adaptive and applicable in any survey based research irrespective of the discipline,
though minor adjustments or variations may be necessary as the case may be.