I'm going to run SEM, as per a journal's reviewer's comment, to explore the relationship between personality traits and management orientations. To collect the data, two questionnaires were used. The first one (personality traits) had 87 items and the second one (management orientations) 24 ones. The first questionnaire included 8 main categories (8 main personality types), while the second questionnaire included 2 main categories (2 main management orientations). Now, I'm wondering whether I have to insert all questionnaire items (which require 111 columns) into SPSS or just the main categories (which require 11 columns).
Thank you very much in advance for helping me with this issue.
P.S.: This is the first time I'm going to run SEM using Amos.
Hello,
Because SEM (factor analysis or path analysis) are estimations of the relations among latent variables (unobservable latent variables estimated from observed indicators variables), you have to encode all the questionnaire items and not 10 categories (8 personality traits and 2 management scores). If you just use your 10 scores (considered as observed variables obtained by the sum or the mean of your items), you are computing a classical regression and don't need SEM. You need to have latent variables in your AMOS model.
See: https://stat.utexas.edu/images/SSC/Site/AMOS_Tutorial.pdf
Best regards
CH
Dear Prof. Houssemand,
Thank you so much for your precise and comprehensive answer as well as the link to the tutorial.
Best regards,
Laya
Dear Laya,
You'll do data entry in SPSS first and then structure the path model (for estimates and fit indices calculations in Amos) after linking the SPSS dataset to the Amos file.
See:
Article When empathy hurts: Modelling university students’ word of m...
Article E-mail Advertising in Syria: Assessing Beliefs, Attitudes, a...
Article Linking information motivation to attitudes towards Web advertising
Article Relating Patient Satisfaction to Nurses’ Job Satisfaction, J...
Dear Dr. Mahmoud,
Thank you very much for sharing the articles. I will read them and they will definitely help me. My first problem was about inserting the data into SPSS. Dr. Houssemand told me to code the items of the questionnaire first and insert all the items into SPSS.
Best regards,
Laya
Concerning the software, I have a preference for R statistical opensource tool. With the LAVAAN package, you can code items in several formats (txt, dat, csv,...) and, after, structure easily your structural model. But the most important option is the possibility to compute your model with the real measurement level of your data. Generally, items from scales (personality, interests,...) have ordinal level of answer (e.g. type Likert). Using the traditional models (e.g. maximum likelihood) you compute the fit of your model as your data were metric. It is not the case. With Lavaan, you can compute that fit using an ordinal calculation. Often, results are really better.
Best regards
Dear Prof. Houssemand,
Thank you for introducing Lavaan-- I didn't know it. It is definitely better than Amos considering the type of data I have. However, I couldn't have found a step-by-step visual tutorial yet and I hope I can find one so that I can use it. I'm not an expert in doing statistics, but I do my best to learn the concepts and how to use Lavaan.
Warmest regards,
Hello,
Thank you so much for your care and consideration.
Best,
Laya
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling