Quantitative methods
This Lab-assignment aims at getting used to SPSS. It emphasizes data management, in addition to analysis of multiple regression, logit regression and discriminant analysis.
The data used for this lab is compiled by you, using data from a data base. You can use firm data provided by the library (Holdings, Infotorg företag etc) or you can use external national (SCB) or international (Eurostat, SCB, OECD or IMF etc) data bases. You should then download/export your data into SPSS and prepare it for analysis.
Lab 1 requires reading of chapters 1, 2, 5,7 and 8 in Hair et al (2018), in particular the applied parts of the chapters. Some of the instructions are implicitly pointed out by the reading instead of explicit lab instructions.
It is essential that you describe what you are doing and why. Just providing output tables from SPSS without showing understanding of what you are doing is not what this exercise is about. It is also essential that you show that you have taken part of the literature via your explanatory texts and the use of references.
1) Data collection
Download data and prepare the data for analysis.
Download data from a data base (see above) and export it to SPSS. The data should include a variety of variables (minimum 5 and maximum 10) and at least 50 observations. All (minimum four) variables should be in ratio scale, except your choice of grouping variables (continents, countries etc). Furthermore, the data should be a cross-section, i.e. from the same year. You are not allowed to share data with your classmates.
Tell SPSS which data level each variable is and name and label the variables. Don’t forget to save the data!
Submission:
- Describe the data and how it was collected (which data base and what the data measures) in text.
- List your variables, its data level (nominal, ordinal, interval, ratio) and the variable label.
2) Data validation
Read Hair et al‘s chapters on Data examination (chapter 2) and Multiple Regression (chapter 5) and apply stage 3 (pp 287-) in chapter 5 to the data. Summarize the data and validate the variables. Describe the methods/tests you are using and what you want to achieve with these in terms of a regression analysis.
Submission:
- Summary statistics relevant for multiple regression analysis.
- Description of data, the methods used for data validation and their results commented.
3) Linear Regression
Formulate a hypothesis (H0/H1) suitable to your data and suitable for a linear regression (a binary test). This is hypothesis 1. Test hypothesis 1 using linear regression in SPSS and interpret the outcomes in terms of the relationship, significance and determination.
Submission
- Hypothesis
- SPSS outcome from a binary linear regression
- Evaluation of essential outcomes
4) Multiple regression
Complement hypothesis 1 with one hypothesis requiring an additional independent ratio variable (Hypothesis 2) and one hypothesis requiring a dummy variable (hypothesis 3). Test the hypotheses using multiple regression analysis in SPSS. Hypothesis 1 should be part of all additional tests and reanalyzed. In addition, you can combine the hypothesis 2 and 3 in the same setting if you want to. Control variables can be used for validation, but should not be a hypothesis. Note that hypothesis 2 and 3 should have the same dependent variable.
Interpret the results and validate the results (consider stages 5 and 6 chapter 5 in Hair et al). Consider all three hypotheses when you evaluate the outcomes.
Submission
- Hypothesis 2 (based on a ratio variable)
- Hypothesis 3 (based on a dummy variable)
- Outcomes of hypothesis 2 and 3 with comments, explanations and analysis.
- Outcomes of hypothesis 1 given hypothesis 2 and 3 (compare outcomes with the bivariate test)
5) Logistic Regression
Read chapter 8 in Hair et al (2018). Use the dependent variable from the regression tests and create a new variable, splitting the variable into two groups, referring to the first hypothesis. Re-do the tests of hypotheses 1-3 using multivariate logistic regression and the new variable as the dependent variable. Analyze the outcomes.
Submission
- Description of how the new variable was constructed (what was the criterion and how it is going to be used analytically).
- SPSS outcomes of the multivariate logistic regression
- Comments and analysis of the hypotheses based on the results of the logistic regression.
6) Discriminant analysis
Read chapter 7 in Hair et al (2018). Use the same data as you did in your logistic regression and do a multiple discriminant analysis. Formulate a hypothesis suitable for multiple discriminant analysis. Test the hypotheses in SPSS and analyze the results.
Submission
- A hypothesis suitable for multiple discriminant analysis including a description of how you can test it using multiple discriminant analysis.
- Outcomes of the multiple discriminant analysis with comments and analyzes.
Review Lab 1
| Name | ||||||
| Not included | Included | Complete | Passlevel | Well done | Very well done | |
| Data Collection | ||||||
| Data Validation | ||||||
| Linear (binary) Regression | ||||||
| Multiple regression | ||||||
| Logistic regression | ||||||
| Discriminant analysis | ||||||
| Use of the literature | ||||||
| Passing grade (Yes/no) | ||||||
| Comment |
