This post is for the basics of Market Research data analysis part. This is not comprehensive but it gives the user the first hand knowledge of how to use SPSS and do basic data analysis, which can be useful for 70% of the cases.
First we need to know more about the data types. These data types are different from the data types which we have in computer language. The data types in this case are of 4 types: Nominal, Ordinal, Interval and Ratio.
Nominal data is when we have to just differentiate between two or more types of data. It is just like assigning a particular code to a specific value. This can be like "Male" and "Female" or say making any specific group of "Users", "Potential Customers" and "Not Interested". This type of data is just choosing a category and so it is also called as categorical data. With this data, we can just do basic mathematical calculations like finding percentage of one type or central tendency (mode) etc. Median and mean doesn't make sense because there is no rank.
Ordinal data is when we have to arrange the data along with a specific name given as in Nominal. It can be like arranging 5 films in a particular order. The films are arranged in an order which shows which is better than the other (in my terms) but it doesn't tell that a film is better by what margin. It shows comparison but not the relative comparison between the two. Here we can have median and percentage. Mean doesn't hold good here too.
First we need to know more about the data types. These data types are different from the data types which we have in computer language. The data types in this case are of 4 types: Nominal, Ordinal, Interval and Ratio.
Nominal data is when we have to just differentiate between two or more types of data. It is just like assigning a particular code to a specific value. This can be like "Male" and "Female" or say making any specific group of "Users", "Potential Customers" and "Not Interested". This type of data is just choosing a category and so it is also called as categorical data. With this data, we can just do basic mathematical calculations like finding percentage of one type or central tendency (mode) etc. Median and mean doesn't make sense because there is no rank.
Ordinal data is when we have to arrange the data along with a specific name given as in Nominal. It can be like arranging 5 films in a particular order. The films are arranged in an order which shows which is better than the other (in my terms) but it doesn't tell that a film is better by what margin. It shows comparison but not the relative comparison between the two. Here we can have median and percentage. Mean doesn't hold good here too.
Interval scale is an ordinal scale which also shows the differentiation between the two data sets. It can be like Celsius Scale of temperature, it shows that 45 degree is 2 degrees more than 43 degrees, but at the same time it doesn't have an absolute zero. Ratios are not allowed since 20 °C cannot be said to be "twice as hot" as 10 °C, nor can multiplication/division be carried out between any two dates directly. We can have mean, median and mode as the characteristics of data.
In case of ratio scale, there is absolute zero defined and so we can have the ratio among the different values. So, weight, age, height etc. all are ratio scales. These are the most mature of all the scales and all the mathematical calculations are possible on this data.
Based on this specific general case of data, we can go to the next level and start using SPSS.
