comments and constructive criticism
The writer will have to read each of this post and react to them by
commenting, analyzing and supporting his reactions with relevant articles.
The writer will have to read carefully before giving constructive comments on
the article. The writer should write a one paragraph of at least 150 words.
APA and in text citation must be use as each respond to the two post must
have in text citations. The writer will have to use an article to supports his
comments in each of the article. Address the content of each post below in a
one paragraph each, analysis and evaluation of the topic, as well as the
integration of relevant resources.
To predict behavior, researchers need to measure behavior. Because it is difficult to observe
behavior, a reliable measurement is difficult. Reliability and Item analysis allow researchers to
construct scales of measurement, improve existing scales and to evaluate scale reliability. A
reliability assessment is based on the correlation between the items that define the scale (Green
(Schriesheim& Kerr, 1974) Schriesheim& Kerr (1974) assessed the validity, reliability
and scaling of different versions of the Ohio State leadership scales; leadership opinion
Questionnaire (LOQ), Leader behavior description questionnaire (LPDQ), and supervisory
behavior description questionnaire (SBDQ). Their assessment identified a few flaws in the
validity of these scales. All of these scales have a high response skewness and have high
correlations. The instruments show issues with scaling. A sufficient number of reflected structure
items do not exist. The response intervals show that the Ohio State leadership scales produce
ordinal data instead of interval data. Internal consistency tests indicate that these leadership
scales meet acceptable reliability standards (Schriesheim& Kerr, 1974).
Current research in leadership does not adequately explain the leadership potential of
college students. Early attempts by Karnes and Chauvin (1985) identified some criteria to
evaluate leadership potential. Although these areas are a good starting point further research is a
need to identify all areas of leadership potential in college students. S. Lee et al. (2015)
developed a measurement scale for leadership potential in Korean undergraduate students. A
random group of 13 leaders with at least five years’ experience were used to test sample
questions. The data obtained was tested to determine if the psychometric qualities of the
proposed leadership potential scale. A confirmatory factor analysis confirmed construct validity.
As a result, of the statistical analysis, a 12 factor model for evaluating leadership potential in
Korean undergraduate students was produced. Although the scale passed initial validity, further
tests are needed to confirm reliability (S. Lee et al., 2015).
Green, S. B., &Salkind, N. J. (2014). Using SPSS for windows and macintosh (7th ed.). Upper
Saddle River, NJ: Pearson.
Lee, S., Kim, H., Park, S., Lee, S., & Yu, J. (2015). Preliminary development of a scale to
measure leadership potentia. Psychological Reports, 117, 51-71.
Schriesheim, C., & Kerr, S. (1974). Psychometric properties of the Ohio State leadership scales.
Psychological Bulletin, 81(11), 756-765.
Include the one paragraph comments hear using and a pear review article to support your
comments. Also include in text citations in APA.
It is indeed important for researcher to develop appropriate measurement scales that are
reliable and valid. Measuring behaviors is one way of helping an individual to understand
the potential of a person. I concur with this observation that sometimes it is difficult to
come up with reliable measuring scales. Nevertheless, the fact that different scales such as
those for measuring leadership exist, it points out that, there is still a room for
improvement (Schriesheim & Kerr, 1974). This as well points out that the available tools
may not be reliable per se. These flaws are manifest in many of the Ohio leadership scales
that have some flaws. On the same note, the article has acknowledged that the scales may
present other flaws such as skewness, high correlations, and validity issues. Therefore, it is
very important that researchers invest their time to test these scales for them to rely on
them. In general, the article is precise and concise. It is objective as it has incorporated or
supported the arguments with references.
Correlation provides a “unitless” measure of association between two variables, ranging from −1
(indicating perfect negative association) to zero (no association) to one (perfect positive
association). Both variables are treated equally in that neither is considered a predictor nor an
outcome (Crawford, 2006). Regression is particularly useful to understand the predictive power
of the independent variables on the dependent variable once a causal relationship has been
confirmed (O’Brien, & Scott, 2012). According to O’Brian and Scott (2012), regression helps a
researcher understand to what extent the change of the value of the dependent variable causes the
change in the value of the independent variables, while other independent variables are held
unchanged. Regression seeks to show how a fixed variable predicts values of the random
values. In a correlation, both variables are random and the researcher shows how they change in
relation to one another. Both procedures seek to show a relationship between variables (Cohen,
Cohen, West, & Aiken 2013). Neither regression nor correlation analyses can be interpreted as
establishing cause-and-effect relationships. They can indicate only how or to what extent
variables are associated with each other. The correlation coefficient measures only the degree of
linear association between two variables (Wiley, 2015). The main difference between
correlation and regression is that in correlation, you sample both measurement variables
randomly from a population, while in regression you choose the values of the independent (X)
variable. One would use regression if you determine the X values before you do the experiment
Bivariate analysis involves the analysis of two variables (often denoted as X, Y), for the purpose
of determining the empirical relationship between them. Bivariate analysis can be helpful in
testing simple hypotheses of association. Bivariate analysis can help determine to what extent it
becomes easier to know and predict a value for one variable if we know the value of the other
variable (Babbie, 2009). Since using a bivariate analysis is best when dealing with multiple
variable, it would be best to use this in my study since I will have multiple variables to test.
Earl R. Babbie (2009). The Practice of Social Research, 12th edition, Wadsworth Publishing, pp.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2013). Applied multiple
regression/correlation analysis for the behavioral sciences. Routledge.
Crawford, S. L. (2006). Correlation and regression. Circulation, 114(19), 2083-2088.
McDonald, J.H. (2014). Handbook of Biological Statistics (3rd ed.). Sparky House Publishing,
Baltimore, Maryland pp 190-208.
O’Brien, D., & Sharkey Scott, P. (2012). Correlation and Regression.
Wiley, J. F., & Pace, L. A. (2015). Correlation and Regression. In Beginning R (pp. 121-137).
Include the one paragraph comments hear using an a pear review article to support your
comments. Also include in text citations in APA.
Understanding different concepts in statistics is expected to be in a position to analyze any
given data. Researchers have to understand the relationships in their data sets, and how to
analyse them statistically. This is also important as it guides them when making decisions
on the best approach to test their information. It is important to differentiate between
regressions, correlation, correlation coefficient and others terms such as bivariate analysis.
These concepts have been clearly discussed in the article and it is easy to understand them.
Any layperson with less knowledge in statistical analysis, can understand the concepts
easily by reading the article. The article has as well made clear definition and pointed out
the differences between various concepts. Furthermore, the article is well formatted and is
supported with enough sources. I also concur that using bivariate analysis is preferred in a
situation where a researcher intend to analyze multiple variables (Babbie, 2009).