Table of Contents
Data is the most important part of scientific research; that allows researchers to understand the concepts necessary to solve a research problem. However, raw data cannot give complete information until it is analysed and organised using a standard method of analysis. Data can be analysed in various ways, and this article will help you get all the essential information about a similar method of analysis named ‘Factor Analysis’. This article will primarily focus on the steps of conducting factor analysis in the dissertation.
Factor Analysis In Dissertation- A Brief Introduction:
Factor analysis is a data reduction process that enables researchers to investigate concepts that cannot be directly measured. It allows researchers to reduce large data sets to provide a handful of comprehensive data that is easy to understand and manipulate in a study. It statistically simplifies the set of complex variables to explore underlying dimensions. Moreover, such dimensions give useful insights into the possible relationship between various variables and factors involved in a phenomenon. Thereby, we can say that the factors analysis statistical method reduces large data sets into small manageable factors so that researchers can deduce useful consequences about the relationship between different variables involved in a study.
Type Of Factor Analysis:
Unlike other types of analysis, factor analysis is used to interpret and analyse data to identify the relationship between variables. Still, instead of grouping responses, it segregates groups as per their co-relevance. Before using factor analysis in the dissertation, students must know about its different types. Factor analysis in a dissertation can be performed in three main types:
Exploratory Factor Analysis:
It belongs to one of the families of multivariate statistical methods that aim to find the smallest number of hypothetical constructs. Hypothetical constructs may include factors, dimensions, synthetic variables, latent variables, and internal attributes. All these constructs explain the variation observed among the data set. This analysis method can be performed using two common methods: R-type-factor analysis and Q-type-factor analysis. The former work on the principle of ‘when factors or variables are calculated from the correlation matrix’, and the latter explores ‘when factors or variables are calculated from the independent respondents’.
You May Also Like: Explanatory Research – Definition, Types & Examples
Confirmatory Factor Analysis:
It is one of the most commonly used methods in social sciences. It helps in testing whether the measure of the construct is consistent with the researchers’ understanding of the concept. Furthermore, it measures whether data fits a hypothesized measurement model. Thus, you must formulate a hypothesis to use confirmatory factor analysis in a dissertation. Using this hypothesis, you may impose constraints that will help to keep the model consistent with theories.
Structural Equation Modelling (SEM):
It is also a multivariate technique often used in a scientific investigation to test and evaluate the causal relationship between factors and variables involved in a study. SEM works by directly measuring the causality among variables based on pre-defined variables. You can do it in various ways, such as by using Latent Growth Modeling, Path Analysis, Confirmatory Composite Analysis, Confirmatory Factor Analysis, and Partial Least Squares Path Modeling.
Seven Steps To Conduct Factor Analysis:
Generally, the steps involved in performing factor analysis in a dissertation are as follows:
- Choosing and measuring variables
- Running correlational matrix
- Pulling out a set of factors
- Specifying the number of factors observed in the correlational matrix
- Rotating the factors
- Interpretation the results
- Reporting the results
After overviewing the steps to conducting factor analysis, many of you may feel overwhelmed. Right? It is a bit tricky to conduct, but it is not draining. Several electronic statistical data analysis packages can share your analysis burden and make the process at least 10X simple for you. Let us briefly see how Statistical Package for Social Sciences Students (SPSS) can help you perform factor analysis in a dissertation.
Also Read About: Semiotic Analysis – Definition, Examples And Uses In Phd Thesis
A Quick Guide To Using Spss For Factor Analysis:
Like other types of statistical analysis in SPSS, factor analysis must also start by entering data into the SPSS spreadsheet. While entering data, you must remember one thing: each participant’s data must be entered in one row. Once you feel confident that all information is entered properly, you must click ‘Analyse’ on the top ribbon. It will open a drop-off menu where you must click ‘Dimension Reduction’ and select ‘Factors’. It will open a factor analysis dialogue box to ask which variables you want to relate.
After specifying the variables, click on ‘Descriptive’ and make a correlational matrix by clicking the ‘Coefficient’ option. Additionally, to test whether the sample size is accurate and if there is enough correlation between variables, you must select the Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test of Sphericity and click on the ‘Continue’ and ‘Extraction’ options. Later, you must click on the ‘Screen plot’ and ‘Unrotated Factor Solution’ by keeping the value of ‘Eigenvalues greater than 1’ in the dialogue box named ‘ Factor Analysis: Extraction’. You can also click ‘Maximum iterations for convergence’ (keep it 25) and ‘Rotation’ (choose Varimax in the rotation dialogue box). You can also select other options only if you want to explore more.
The above steps will help you conduct factor analysis in a dissertation by following only a few simple steps. These steps will also save you from applying mathematical formulae at each step. The graphs you create using SPSS are easy to interpret and help you reach valid conclusions quickly. If you still feel confused, you must seek help from data analysis experts at dissertation writing services UK.
Final Thoughts:
All in all, data analysis by factor analysis in a dissertation facilitates researchers in solving many scientific queries, especially for finding a causal relationship between different factors. To make valid inferences, you must know the basic aims and objectives of conducting this analysis. You can perform it manually by following seven simple steps, but by using an automated statistical analysis tool, you can perform the task by clicking only a few important options.
