Consider the following questions asked by a researcher in a questionnaire on Social Acceptance: The product should be approved by my friends. The product should be popular in my circle of friends. The product should make a good impression on people who are important to me. The product should be liked by people who are […]
According to Gefen and Straub (2005), “discriminant validity is shown when each measurement item correlates weakly with another construct excepts for the ones to which it is theoretically associated”. In cross-loadings, the researcher examines the various items to identify those that have high loadings on the same construct and those that load highly on multiple […]
Because constructs are measured indirectly, it is recommended that you have at least three items/indicators/questions measuring each of them. For example, in the data used for this post, university faculty members were asked about their perceived quality of information in Wikipedia using the following five different questions. QU1: Articles in Wikipedia are reliable QU2: Articles […]
The table below presents two different tests: the Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy and Bartlett’s test of Sphericity. KMO KMO is a test conducted to examine the strength of the partial correlation (how the factors explain each other) between the variables. When we talk about communalities, we are interested in the amount of original […]
In Exploratory Factor Analysis (EFA), the researcher has no prior intention of confirming the existence of any existing relationship prior to the analysis of the data, but instead, the researcher allows the various techniques, methods, procedures and rules to reveal the relationship in the data. EFA is indeed the opposite of Confirmatory Factor Analysis (CFA) […]
Convergent validity is a subset of construct validity. Here, the researcher’s aim is to find out whether the items he claims are measuring a particular construct are indeed measuring them. In other words, if these items do measure a specific construct, then they need to converge. The researcher achieves this by taking into consideration All […]
Applied research are those research that are meant to directly solve problems to improve human conditions. An example is the research conducted to produce vaccine for the deadly coronavirus. Why? Because scientist want to eradicate the virus from the earth’s surface, hence saving lives. Fundamental research is a kind of research solely meant for knowledge […]
Exogenous variables are also known as predictors/independent variables. They are those variables that cause changes in other variables (endogenous variable). Changes that occur in the exogenous variables are as a result of respondents’ demographical characteristics and hence cannot be explained by the model.
Your main reason for conducting discriminant validity for your study will be to show how distinct an item or set of items is from others. In other words, you are interested in showing that items measuring different constructs or variables have poor relationships or low correlation exist between them. If you fail to test for […]
Common method bias is normally prevalent in studies where data for both independent and dependent variables are obtained from the same person in the same measurement context using the same item context and similar item characteristics. Sources of common method bias Independent and dependent variables are used with the same item. The presence of errors […]
In the model summary of your regression output, you see values of R, R Square, Adjusted R Square, R Square Change and F Change. This post will teach you the right way of interpretating them with good examples.
You can’t trust outliers and their effect maybe be minimal or larger than you think. Their presence can result in a false positive or nagtive result. Let me walk you through the fundamental steps to address this issue using box and whisker plot.
The sole aim of this tutorial is to teach you how to assess common method bias using Harman’s one-factor test for your study especially when the same measurement instrument is used to collect data for both dependent and independent variables.
P-value is also known as probability value and it ranges from 0 to 1. The interpretation of a p-value varies depending on what one is studying. For instance, while a small p-value less than 0.05 provide good evidence against the null hypotheses for ANOVA, t-test, Pearson’s correlation, regression analysis, etc., in some cases such as […]
