Enhanced Concepts on Data types with Special Reference to Questionnaires
The conceptual understanding of the nature of data is very essential in any research design. The data can be of Nominal, Ordinal and Scale type. Though SPSS automatically deals the data around this but at the conceptual level, it is quite essential to understand the backdrop. Nominal data are the data for the namesake. One can only count this type of data. Having no prior information on these types of data given you almost no information before hand. No arithmetic operations can be performed upon them. Ordinal data stands for an order, for e.g. you find the data around the education background in the order of high school, inter, U.G, P.G, doctoral and post-doctoral. The sequence of these data is fixed. If only a degree is named say U.G, one doesn’t make any sense out of it unless proper information on the background is provided.
Scale data on the other hand gives an added comprehension of the dataset. For e.g. comparison of monthly shopping of the grocery of any two households can be understood by the net amount spent by each family. The net amount here also gives us extent of the difference between the two households spending. Thus this enables us to comment on the consumption level of each family.
Scale data becomes quite essential for conducting regression, correlation, scatter plot etc. Scale data can also be converted into nominal/ordinal data by making groups and dividing them into categories.
When preparing the questionnaires, the nature of the data should be taken care of. Another aspect in the questionnaire design is to provide the scope for the missing values if the data type suggests the same (context: SPSS). As many questions are there which the respondents would not be able to answer or never like to answer. Missing values are not taken as values; the presence of missing value indicates that the data is of scale type. As this indicates that the data can be endless. On the other hand, wherever you find a value, it becomes a category; as it puts a limit to the flow of the data. Thus such data can belong to nominal/ordinal type of data.
Section B Group 6_Vivek Roy (12FPM005)
- Apurva Ramteke(13PGP068)
- Chandan Parsad(13FPM002)
- Komal Suchak (13PGP086)
- Rohan Kr. Jha (13FPM004)
- Silpa Bahera (13PGP107)
- Sushil Kumar (13FPM010)
- Vaneet Bhatia (13FPM008)