Ch 13Organisation of Data — Class 11 Economics, concept-first.

Once data have been collected — whether by census or by sampling — the next step is to organise them. Freshly gathered data are usually raw: disorganised, often very large, and cumbersome to handle.

Data as first collected are raw or unclassified — disorganised, usually bulky, and awkward to work with.

Raw data can be grouped in several ways, and the choice depends on the purpose — just as books could be classified subject-wise, author-wise (alphabetically) or by year of publication.

A variable can be measured, but variables differ in how they vary. On this basis they fall into two types.

A frequency distribution is a full way of classifying the raw data of a quantitative variable. It shows how the different values of the variable are spread across several classes, together with each c…

Building a frequency distribution from raw data comes down to answering five questions, each taken up in turn in the sub-sections that follow:

Class intervals are usually kept equal, but unequal intervals are the better choice in two situations.

The number of classes is normally kept between six and fifteen — few enough to summarise, many enough to preserve detail.

The size of each class and the number of classes are two sides of the same decision — neither can be fixed without the other.

Class limits must be definite and clearly stated. Open-ended classes such as "70 and over" or "less than 10" are generally avoided.

Take marks in a test recorded only as whole numbers (no fractional marks), ranging from 0 to 100 — a discrete variable.

For a continuous variable such as height (cm) or weight (kg), the inclusive-looking classes are understood to stretch right up to — but not including — the next lower limit.

When a continuous variable is classified by the inclusive method, as in Table 3.4 above, the classes leave visible gaps.

Frequency of a value is simply how many times it appears in the raw data. In the 100 mathematics marks, 40 occurs 3 times, 0 occurs once, 10 occurs once, 49 occurs 5 times, and so on — so the frequenc…

Tally marking is the practical way to count how many observations fall in each class. Go through the raw data one value at a time and place a stroke ( / ) against the class it belongs to: a mark of 57…

Grouping raw data into a frequency distribution has one built-in drawback: a loss of information. The table is concise and easy to read, but the individual observations disappear.

Equal class intervals are not always best. In the mathematics-marks distribution most observations pile into just three classes — 40–50, 50–60 and 60–70, with frequencies 21, 23 and 19.

Everything so far grouped a continuous variable into class intervals. For a discrete variable the corresponding arrangement is a frequency array.

Everything so far dealt with one variable at a time (a univariate distribution). But a sample often records two pieces of information per unit — for example, from 20 firms both their sales and their a…

Data from primary and secondary sources arrive raw and unclassified, so the step after collection is to classify them for analysis — because classification brings order to what is otherwise unmanageab…