In statistical researches, a description of data and analysis of a hypothesis is so crucial.
Additionally, a keen assessment of relationships that occur amongst variables is essential since
this will lead to an accurate solution to the study problem. Determining the causes and the effects
of variance that exists in variables is essential in making informed changes. This paper will focus
on the description of various aspects of statistics in research.
Describing data is essential in research since it shows a summary of patterns that emerge
from a given set of data. For simple interpretation of data, it is vital to illustrate the measures of
central tendency by determining the median, mean, and mode. Also, one should describe how
spread out the data is by calculating the various measures of spread such as variance and
standard deviation. It is important to summarise data using tables, graphs, charts, and
arithmetical commentary (Fundi et al., 2017; p.1-6). In the natural resource industry, describing
data has enhanced predictive modeling to aid in making a decision that has been applied in
ingesting and integrating large amounts of data from graphical, geospatial text, and temporal
data.
Rees and David (2018; pp139-60) explains that testing a hypothesis enables the
experimenter to determine if the observation of a given phenomenon is likely to have actually
occurred established on statistics. Additionally, it unravels the differences in groups, associations
between variables, and the effects of certain treatments. For instance, if one rejects a null
hypothesis, it is proof that the outcome is statistically significant; therefore, coincidental and luck
played no role. Contrary, if one fails to reject a null assumption, then it implies that the study had
no influence or difference. Medical procedures and pharmaceutical drugs are tested using this
method.
STATISTICS IN RESEARCH 3
Examining a relationship between variables provides a platform for regression that uses
the proven correlations between dependent and independent variables to predict the values of the
dependent variable. Additionally, correlation matrices play an important role in obtaining factor
solutions by studying the construct validity of data (Losh & Susan, 2017; p.1-3). Assessment of
relationship is key during the development and testing of theoretical models by precisely
explaining the nature of bivariate correlations.
Oyediran et al. (2018; p.1-6) defined a research problem as an exact difficulty,
contradiction, issue, or gap in information that one focuses on. There are two types of research
problems. First are practical problems that contribute to change. For instance, you can look for
processes that could be improved in an organization. Second is theoretical research problems
aimed at expanding knowledge or understanding — for instance, investigation on the connection
between economic growth, population size, and political status. Mostly the aim of the research
should be adhered to throughout the process to realize accurate outcomes.
According to Khan et al. (2019; p. 7-11), causal-comparative is an attempt to find out the
consequences or cause of differences among groups of individuals. There exist three types of
Causal comparative research; exploration of effects, cause, and consequences. Auditorium
testing refers to allowing the target audience to rate your research by use of various ways,
including electronic gadgets or questionnaires. Treatment groups are test subjects and variables
under study. Analogous is a degree of similarity between two variables. Pseudo-experimental
refers to using inferential statistics to ascertain treatment effects with an error term inappropriate
to the hypothesis under consideration. It is important to include the control in any experiment for
clarity in comparison of results.
STATISTICS IN RESEARCH 4
In summary, the choice of an appropriate research design is crucial in attaining accurate
results for any study problems. All the variables under analysis should be analyzed by the use of
relevant tests to unravel their relationships. The most significant part of any research is the
ability to present the results clearly for easy understanding.
STATISTICS IN RESEARCH 5
References
Fundi, M., Clements, L., Stern, D., Stern, R., Renaud, F., & Sananka, A. (2017). Describing Data
Well in R-Instat. In Proceedings of the 2017 IASE Satellite Conference, Rabat, Morocco.
Revised from;http://www.africanmathsinitiative.net/wp-
content/uploads/2017/01/Describing-data-well-in-R-Instat.pdf
Khan, A. M., & Ramzan, A. (2019). Casual Comparative Investigation of J. C Maxwell's
Leadership Levels and its Impact on Organizational Change at Elementary School
Level. Paradigms, 13(1), 7-11. Revised from; A Ramzan – Paradigms, 2019 –
search.proquest.com
Losh, S. C. (2017). Dependent and Independent Variables. The Wiley‐Blackwell Encyclopedia
of Social Theory, 1-3. Revised from;
https://onlinelibrary.wiley.com/doi/pdf/10.1002/9781118430873.est0622
Oyediran, K. K., & Oyediran, A. J. (2018). Teaching Basic Terms in Statistics through
Storytelling and Linkages. The Journal of Middle East and North Africa
Sciences, 10(5777), 1-6.Revised from; https://www.jomenas.org/
Rees, D. G. (2018). Essential statistics. Chapman and Hall/CRC. Revised from; DG Rees –
2018 – content.taylorfrancis.com