Page 152 - Data Science class 11
P. 152

randoMisation




                                                                                    04










                    Learning Outcome


                  4.1  Data
                  4.2  Some Statistical Terms
                  4.3  Surveys
                  4.4  Sampling
                  4.5  Level of Confidence: How sure are you?
                  4.6  Automatic Data Collection via In-built Sensors
                  4.7  Structuring Data using XML
                  4.8  Structure of an XML Document



        In the previous chapter, you learnt about how one can use observational studies to collect data. The collected data
        is further analysed to derive a conclusion. In this chapter, you will learn about how one can collect data via various
        media like surveys, sensory devices, and the Internet. You will also explore a way to increase the accuracy of the results
        deduced using a confidence interval.


        4.1 Data

        As we already know, data is the foundation of data science. All the analyses are based on the type of data. Therefore,
        it is a must  to understand the different types of data.

        4.1.1 types of Data
        Let us now learn about the two types of data.

        Primary Data
        Primary data is a type of data that is collected directly from first-hand sources like interviews, surveys, experiments,
        etc. by researchers. Primary data is generally collected from the source—where the data initially originates from. This
        data is considered the premier data in research.
        The Advantages of Primary Data

           • Specific: Primary data is particular to the needs of the researcher during data collection. The researcher is able to
          control the type of data that is being collected.
           • Controllable: The researcher has complete control over the data gathered through primary research. He can decide
          which design, method, and data analysis techniques should be used.
           • Accurate: It is more precise than secondary data. The data is free from personal bias and is authentic.



          150   Touchpad Data Science-XI
   147   148   149   150   151   152   153   154   155   156   157