Summary
Highlights
A good classification must be comprehensive (no item left out), clear (unambiguous grouping), homogeneous (similar items in a class), suitable (aligned with research purpose), stable (consistent classification basis), and elastic (allowing for minor adjustments).
Data can be classified into four types: Geographical (based on location), Chronological (based on time), Qualitative (based on attributes like rich/poor or educated/uneducated, further divided into simple and manifold), and Quantitative (based on numerical values like salary or marks).
Organisation of data is the second stage of statistical study after data collection. It involves systematically arranging numerical data to enable comparison, analysis, and drawing conclusions. Classification is the best way to organize data, which means dividing data into groups based on characteristics like academic streams or marks obtained in a test.
The main objectives of classification are to make data brief and simple, increase its utility, highlight distinctions, facilitate comparison, provide a scientific arrangement, and enhance attractiveness and effectiveness for a lasting impression.
A variable is a measurable characteristic whose value changes over time. Variables are of two types: Discrete (values increase in complete numbers or jump, cannot be in fractions or decimals, e.g., number of students) and Continuous (values increase in fractions or decimals, or are presented in ranges, e.g., height or weight).
Raw data is unorganized, crude data. When raw data is arranged in a systematic order (ascending, descending, or serial number), it forms a statistical series. There are three types of series: Individual Series (items listed singly without frequency), Discrete Series (items with their frequencies), and Frequency Distribution Series (items grouped into class intervals with their frequencies).
Key terms include: Frequency (number of times an item repeats), Class Frequency (frequency for a class interval), Tele Bars (graphical representation of frequency, using 'four and cross' method), Class (the groups or ranges, e.g., 0-10, 10-20), Class Limits (lower and upper limits of a class, denoted as L1 and L2 respectively), Size of Class Interval (L2 - L1), Range (maximum value - minimum value), and Mid-value ( (L1 + L2) / 2 ).
Frequency distribution series are categorized into five types: Exclusive Series (upper limit of a class is excluded and included in the next class), Inclusive Series (both lower and upper limits are included within the same class, can be converted to exclusive by adjusting 0.5), Open-End Series (first class has no lower limit or last class has no upper limit), Cumulative Frequency Series (frequencies are continuously added, creating 'less than' or 'more than' series), and Mid-Value Frequency Series (class intervals replaced by their mid-values, which can be converted back to class intervals using formulas involving mid-value and class width).