Data Collection Classification Presentation Questions for SSC CGL

Types of Data: Primary vs Secondary

Primary data is collected directly by the researcher through surveys, interviews, or experiments. Secondary data is already collected and published by someone else — you just use it. In SSC exams, you'll see questions asking you to identify which type of data a scenario describes. Primary data is original but time-consuming; secondary data is quick but may be outdated or biased by the original collector's method.

• Primary data: collected firsthand, original, reliable for your specific need, expensive and time-consuming
• Secondary data: already exists (books, reports, websites), cheaper, faster, but may not fit your exact requirement
• Primary examples: survey, census, interview, observation, experiment
• Secondary examples: published reports, government statistics, newspaper articles, research papers
• SSC trick: if the question says 'data from a government website' or 'from a published report', it's secondary

Classification of Data: Qualitative vs Quantitative

Quantitative data is numerical and measurable (height, age, salary). Qualitative data is descriptive and categorical (color, opinion, gender). SSC questions often ask you to classify a given variable. Remember: if you can count or measure it with numbers, it's quantitative; if it describes a quality or category, it's qualitative.

• Quantitative: numerical, can be added/averaged, examples are height, weight, income, marks
• Qualitative: descriptive, categorical, examples are color, religion, satisfaction level, blood type
• Discrete quantitative: whole numbers only (number of students, number of cars)
• Continuous quantitative: any value in a range (temperature, weight, time)
• SSC pattern: expect 'classify the following' questions mixing both types

Sampling Methods

When you cannot survey everyone (population), you select a sample. SSC tests your knowledge of common sampling techniques. Random sampling gives every unit equal chance; stratified sampling divides population into groups first; systematic sampling picks every nth item; cluster sampling divides into clusters and randomly picks clusters. Convenience sampling is non-random and biased — avoid it in real studies.

• Random sampling: every unit has equal probability, unbiased, most reliable
• Stratified sampling: divide population into strata (layers), then randomly sample from each stratum — used when population is diverse
• Systematic sampling: select every kth item from a list (e.g., every 5th person) — simple and practical
• Cluster sampling: divide into clusters, randomly select clusters, survey all units in chosen clusters
• Convenience sampling: pick easily available units, biased, not recommended for research
• SSC asks: 'Which method ensures every unit has equal chance?' Answer: Random

Data Presentation: Tables, Graphs and Charts

Raw data is hard to understand. Presentation makes it clear. Frequency tables organize data by categories and counts. Bar charts compare categories (use for qualitative or discrete data). Histograms show distribution of continuous data (bars touch each other). Pie charts show parts of a whole as percentages. Line graphs show trends over time. SSC Tier-II often includes a graph and asks you to read or interpret it.

• Frequency table: lists categories and their frequencies (counts), helps identify mode and patterns
• Bar chart: vertical or horizontal bars for categorical data, bars do NOT touch
• Histogram: bars for continuous data, bars TOUCH each other, shows distribution shape
• Pie chart: circle divided into slices, each slice is a percentage of total, shows composition
• Line graph: points connected by lines, shows trend over time, ideal for time-series data
• SSC tip: if question shows a graph, read axis labels carefully and check the scale

Frequency Distribution and Grouped Data

When data has many values, you group them into classes (intervals). A frequency distribution table shows each class and how many data points fall in it. Class width is the range of each interval. Class mark (midpoint) is used in calculations. Cumulative frequency adds up frequencies as you go down the table. SSC may ask you to find median or mode from a frequency table.

• Class interval: a range like 10-20, 20-30 (usually written as 10-19, 20-29 to avoid overlap)
• Class width: difference between upper and lower limits of a class
• Class mark (midpoint): (lower limit + upper limit) / 2, used in mean calculation
• Frequency: count of data points in each class
• Cumulative frequency: running total of frequencies, used to find median
• SSC pattern: given a frequency table, find mean, median, or mode

Measures of Central Tendency from Grouped Data

For grouped data, you cannot calculate mean, median, or mode exactly — only estimate. Mean is calculated using class marks and frequencies. Median is found using cumulative frequency and the formula. Mode is the class with highest frequency (modal class). SSC Tier-II includes 1-2 questions on this, usually medium difficulty.

• Mean of grouped data: sum of (class mark × frequency) divided by total frequency
• Median of grouped data: use cumulative frequency to find the class containing the median, then apply formula
• Mode of grouped data: the class with the highest frequency is the modal class
• These are estimates, not exact values, because original data is lost in grouping
• Always check that total frequency equals number of data points

• Confusing primary and secondary data: remember, primary is collected by YOU, secondary is already collected by someone else. If it says 'from a published report', it's secondary.
• Mixing up bar chart and histogram: bar charts have gaps between bars (for categories), histograms have touching bars (for continuous data ranges).
• Forgetting class width in grouped data: when forming class intervals, ensure they are equal width and non-overlapping. Common error is writing 10-20, 20-30 (overlap at 20) instead of 10-19, 20-29.
• Using wrong sampling method: random sampling is NOT the same as convenience sampling. SSC often tests whether you know random sampling is unbiased.
• Misreading graph scales: always check the axis labels and scale. A graph showing 0-100 looks different from 0-1000 even if the bars look the same height.

• PRIMARY = Personally Researched, Immediate, Mine. SECONDARY = Someone Else's, Secondhand, Source.
• QUANT = Numbers (Quantitative). QUAL = Qualities/Categories (Qualitative).
• BAR = Gaps (categorical). HISTOGRAM = Hugging (continuous, bars touch).
• RANDOM = Reliable, unbiased. CONVENIENCE = Cheap but Crooked (biased).
• Frequency table → Mean (use class marks). Cumulative frequency → Median (use CF formula). Highest frequency → Mode (modal class).

• SSC Tier-II Quant: expect 1-2 direct questions on data classification (primary vs secondary, qualitative vs quantitative). These are usually easy 1-mark questions to build confidence.
• Graph interpretation: 1-2 questions show a table or graph and ask you to read values, calculate totals, or identify trends. Practice reading axis labels and scales carefully — this is where careless errors happen.
• Grouped data calculations: if a frequency table is given, expect a question on mean or median. The median formula is the trickiest part — memorize the formula and practice 2-3 examples before exam.
• Sampling methods: SSC may ask 'which method ensures every unit has equal chance?' (Answer: random) or 'which method is used when population is diverse?' (Answer: stratified). These are conceptual, not calculation-based.
• Time management: data presentation questions are usually quick if you understand the concept. Spend 2-3 minutes max per question. If a graph question is confusing, skip and come back — don't waste time.