Collecting Quantitative and Qualitative Data

Sampling and Collecting Quantitative and Qualitative Data
Probability sampling represents the best strategy for selecting research participants
There are many ways of gathering information from a population using the qualitative or the quantitative approaches. Probability sampling is indeed the best strategy that a researcher can use to select participants from a population. There are many reasons to support this position as below discussed. Probability sampling can be understood as the sampling technique that takes a small sample from larger populations using one of the methods under the probability theories. In order for a participant to be selected a random method of selection must be used. The most fundamental requirement in probability sampling is ensuring that every person in the population has an equal and known likelihood of being selected. For instance, where a sample of 10 people is to be selected from a population of 100 people every person in that population would have a 1/10 chance of being selected. With probability sampling a researcher has the best opportunity of coming up with a sample size that truly reflects the entire population.
Statistical theory is applied in probability sampling for random sample selection. From the selected sample the researcher predicts that the entire population would have a similar response. The beauty of probability sampling is in its simplicity. Samples are selected in a fashion that allows every member in a population an equal opportunity of selection. With randomization any sampling and systematic bias is alleviated. This essentially means that the selected sample is an accurate representation of the whole population. With probability sampling there is less judgment. The researcher does not influence the selection process in any way. This makes the process more accurate and effective. Probability sampling is simple enough to be executed by non-technical researchers. Probability sampling does not include any long or complex processes. It is quite easy to execute.
Onwuegbuzie & Collins (2007) consider that prior to making a decision on the scheme of sampling it is a requirement for the researchers to decide the study objective. For instance, if the purpose is to generalize qualitative and quantitative research findings to a given population where inferences were made, the researcher should then try to choose a random sample with those components. Such situations will require a mixed method research to consider one of the five random sampling methods at some point during the process of research (Onwuegbuzie & Collins, 2007). The random selection methods include cluster, stratified, simple, multi-stage, and systematic random sampling methods. This means that probability sampling is beneficial for quantitative research as well as mixed method research approaches (Teddlie & Yu, 2007).
Data Collection Method: Interviews
Interviews enable accurate screening because the interviewee is able to capture both the non-verbal and verbal communication. Interviews also help in capturing accurate data with respect to conditions such as race, age and gender. The interviewer is also able to keep the interviewee focused and hence the ability to capture the actual behavior and emotions necessary to make a determination. With interviews the researcher has the ability to choose the right candidate for a research. With an interview the interviewer is also able to capture primary information. The interviewer has the latitude to ask as many questions and clarify issues with the interviewee.
Interviews can be limited by factors such as cost. The process is quite intensive and costly. The skills and abilities of the interviewer determine the quality of the data collected (Byrne, 2001). Interviews can be quite limiting depending on location and size of the team doing the interviewing. Moreover, the questions asked can be deliberately or accidentally biased by the way they are phrased or ordered. It is possible to influence the answers given by the interviewee. The fear of being judged can also cause an interviewee to give a dishonest answer.
Ethical issues in an interview include confidentiality, bias and permissions. The interviewee should always remain anonymous unless with their consent. Identification of the interviewee should only be allowed if it is essential for the accurate conclusion of the research (Byrne, 2001). Bias in an interview can come in many different ways including asking leading questions that influence the answers given. The selection process can also be skewed to favor some candidates. Interviews rely heavily on good judgment. If the judgment of the interviewer is poor then the results will also be biased. It takes a lot of preparation and skill to conduct a successful interview. The information corrected through writing, tape recording or notes must be used in accordance with the interviewee wishes. The interviewee should give a written consent detailing the permissions granted.
Measurement reliability and measurement validity
According to Drost (2011) reliability can be understood as extent to which measurements can be repeatable. Whenever different candidates conduct the measurements in differing occasions, using alternative instruments for the same thing, and under varying circumstances the consistency of the outcome is what defines measurement reliability (Drost, 2011). In summary reliability can be understood as measurement consistency or measurement stability over various conditions where similar results are obtained.
Measurement validity according to Drost (2011) deals with research components meaningfulness. For instance, when a researcher is measuring behavior their concern is whether they indeed are measuring what was purposed. Does the Intelligence Quotient (IQ) measure level of intelligence? Such a question examines measurement validity. Although complete certainty is not guaranteed in these instances a researcher can build a strong foundation for validation of the measures.


References
Byrne, M. (2001). Interviewing as a data collection method. AORN Journal, 74(2), 233–235. https://doi.org/10.1016/S0001-2092 (06)61533-0
Drost, E. A. (2011). Validity and reliability in social science research. Education Research and Perspectives, 38(1), 105–124.
Onwuegbuzie, A. J., & Collins, K. M. (2007). A typology of mixed methods sampling designs in social science research. The Qualitative Report, 12(2), 281–316.
Teddlie, C., & Yu, F. (2007). Mixed Methods Sampling: A Typology with Examples. Journal of Mixed Methods Research, 1(1), 77–77. https://doi.org/10.1177/1558689806292430