Your class wants to pick a class pet. You could guess that most kids want a hamster, or you could ask. Data collection means asking or watching in a planned way and writing down every answer. If 12 kids say hamster, 8 say fish and 5 say turtle, you no longer have to guess.
There are three main ways to collect data. A survey asks people a question, like 'Which pet do you want?'. An observation means watching and counting, like counting birds at a feeder for 15 minutes. An experiment means changing one thing on purpose and measuring what happens, like giving one plant more light than another and measuring both every few days.
How you collect matters as much as what you count. If you ask only your 5 best friends and they all love turtles, you might think turtles win, even though most of the class wants a hamster. A fair plan asks everyone, or a mix picked by chance, and asks every person the same question in the same words.
Not every question needs data collection. 'How many days are in a week?' has one answer you can look up. You collect data when the answers will be different from person to person or from day to day.
Think of it like this. Collecting data is like a shop worker counting what is on the shelves. She doesn't guess; she checks every shelf and writes each item down. Skip a shelf and the list is wrong, even if the counting was perfect. Unlike cans on a shelf, people's answers can change, so data tells you about the day you collected it.
Core idea. To answer a question about a group, plan what to ask or count, ask everyone (or a fair mix) in the same way, and record every answer; the plan decides whether you can trust the result.