Hypotheses (AQA GCSE Statistics): Revision Notes
Hypotheses
What is a hypothesis?
A hypothesis is a clear, precise statement about something you can measure that can be tested by gathering and examining data. Think of it as an educated guess or prediction that you want to investigate. The hypothesis might turn out to be correct or incorrect - that's what makes it worth testing!
The key purpose of a hypothesis is to guide your data collection. You need to gather relevant information and study it carefully to determine whether your hypothesis is supported by the evidence or contradicted by it.
Requirements for a good hypothesis
For a hypothesis to be useful in statistical investigation, it must meet two essential criteria:
Specific
Your hypothesis needs to be precise and clearly defined. Vague statements that can be interpreted in multiple ways won't work for data collection.
Poor example: "Swimming for 5 hours or more a week improves your heart rate"
- This is too vague because "improves" could mean many different things
- It doesn't specify what type of improvement or how it's measured
Measurable
Your hypothesis must involve something you can actually collect numerical data about. You need to be able to record, count, or measure the variables you're investigating.
Good example: "As people get older they need more sleep"
- This works because you can measure age in years and sleep in hours
- Both variables can be recorded as numerical data and grouped appropriately
Writing hypotheses - practical considerations
When crafting your hypothesis, you need to think carefully about how the data collection will work in practice.
Planning your data collection:
Age groups - Consider what age categories make sense for your investigation. For example:
- Less than 30 years
- 30 to 50 years
- Over 50 years
Clear age brackets help you organise your data collection and make meaningful comparisons between groups.
Timeframe for data collection - You must also identify the period over which you'll gather information. For instance, if investigating gym usage:
- How many hours do you spend at the gym on average per week?
- 0 hours
- Fewer than 5 hours
- 5 to 10 hours
- More than 10 hours
Worked examples
Worked Example 1: Gym usage hypothesis
Matt's original hypothesis: "Young people spend more time at the gym than old people."
Problem with this hypothesis: The statement is not precise and not measurable. Terms like "young" and "old" are not clearly defined - what ages count as young or old?
Improved hypothesis: "People under 30 spend more time at the gym than people over 50."
This works better because:
- It specifies exact age boundaries (under 30 vs over 50)
- Time spent at the gym can be measured in hours per week
- Both variables can be collected as numerical data
Worked Example 2: Mobile phone usage hypothesis
Rosa's investigation: She wants to examine the amount of time male and female students spend using their mobile phones each day.
Good hypothesis: "Female students spend more time on their mobile phones than male students."
Data collection needed:
- Primary data: Amount of time spent per day on mobile phones (measured to the nearest minute)
- Categories: Males and females
- This is primary data because Rosa would collect it directly from students
Alternative hypotheses could include:
- "Males spend more time on their mobile phones than females"
- "Females spend the same amount of time on their mobile phones as males"
Common exam tips
Watch out for these traps:
- Don't use vague terms like "young/old", "lots/little", "better/worse" without defining them precisely
- Make sure both variables in your hypothesis can be measured numerically
- Always consider what data you would actually need to collect to test your hypothesis
- Think about whether you're collecting primary data (directly from people) or secondary data (from existing sources)
Problem-solving approach:
- Read the hypothesis carefully
- Identify what needs to be measured
- Check if the terms are specific enough
- Consider how you would collect the data
- Suggest improvements if needed
Key Points to Remember:
-
A hypothesis must be both specific (clearly defined terms) and measurable (numerical data can be collected)
-
Think practically about data collection - what age groups, timeframes, and categories make sense?
-
Avoid vague language - define terms like "young", "old", "better", "more" with precise boundaries
-
Consider whether you need primary data (collected directly) or secondary data (from existing sources)
-
A good hypothesis guides your entire data collection process, so it needs to be crystal clear from the start