Darryl Laws
The construction of Likert scale is rooted into the aim of my research, the purpose of which is to understand about the opinions and perceptions of my elite purposeful selected sample’s (participants) related to my latent variables (phenomenon of interest) irrational behavior. This latent variable is expressed by several manifested questions or statements within my questionnaire (herein). These specifically constructed questions / statements in a mutually exclusive manner address a specific dimension of phenomenon (irrational behavior) under inquiry and in a cohesive measure of the whole phenomena and seek to ascertain whether or not observable measures are driven by the same underlying variable. During analysis, the scores of the all questions / statements of the questionnaire are combined (sum) to generate a composite score, which logically in totality measures anuni-dimensional trait.
Inferential statistics. Explanatory Factor Analysis (“EFA”) and Principal Component Analysis (“PCA” are techniques for identifying clusters of variables (Field, 2018). These techniques have three main uses: 1) to understand the structure of a set of variables, 2) to construct a questionnaire to measure an underlying variable and 3) to reduce a data set to a more manageable size while retain as much of the information as possible (Field, 2018). I will use both “EFA” and “PCA” which are aimed to reduce a set of variables into a smaller set of dimensions called factors and components.
I will use Factor Analysis to explain the maximum amount of common variance in a correlation matrix using the smallest number of explanatory constructs. (Explanatory constructs = latent variables (factors) and they represent the cluster of variables that correlate with each other (Field, 2018). Whereas “PCA” tries to explain the maximum amount of total variance in a correlation matrix by transforming the original variables into linear components (Field, 2018). The Likert scale measurements gives me the foundation in which to integrate my raw data into my models in SPSS Version 26 and run the reliability measures as well as EFA and PCA.
The three primary reasons for use of a Likert Scale are:
No limitation to binary questions. Likert scales offer survey respondents the opportunity to indicate the extent to which they agree or disagree with a given statement or to express a neutral response. People are not forced into making a binary choice between ‘agree’ and ‘disagree’. The quantitative survey data can then be readily collated.
Data Opinions. By way of an example; when using the Likert scale, there is no need to reword the question What do you think of our current President? There is only the need to provide the right choices. For this example:
He is the best President ever.
He is doing pretty well.
He is alright.
I don’t care.
He is not doing okay.
He is doing a terrible job.
He is the worst President ever.
By providing respondents with these choices a surveyor can get more granular feedback.
Comments
Post a Comment