Science

Practical Investigation

Variables are all the things that could change during an investigation.

In a bouncing ball investigation, where the height a ball bounces to is measured after it is dropped at different heights, many things could affect the results from one experiment to the next such as using a different ball, a different drop height or a different surface which the ball is dropped on.

​

You should only change one thing at a time in your investigation. This called the independent variable.(The height the ball is dropped at)

During your investigation you should be able to measure something changing which is called the dependent variable. (How high the ball bounces after being dropped)

The factors you keep the same in your experiments (fair test) are called control variables

A 'fair test' is one in which you only change one thing (variable).

Scientific investigations are typically written up in a standard way under the following headings:

​

Aim (focus question): what you are trying to find out or prove by doing the investigation

Hypothesis: what you think will occur when an investigation is carried out

Equipment (or materials): the things that you need to do the investigation

Method : A simple, workable plan of what you will do – and can be repeated by another person. Usually written as a numbered list

Results : data, tables and graphs collected from investigation

Conclusion : what your results tell you – linked back to the aim and hypothesis

Discussion : Science ideas to explain your results, possible improvements to the investigation, how you managed to control the other variables.

​

The typical way that scientists work is called the Scientific method.

Focus Question / Aim

Your Aim or focus question must include both variables.

For example: If I change (independent variable) how will it affect (dependant variable)

​

Such as: If I change the temperature of the water (independent) how will it affect how much sugar I can dissolve into the water (dependant)

​

Independent variable – amount of light a plant receives

Dependant variable - height that plant grows

​

Focus Question: How does the amount of light a plant receives affect the height it grows to

EXAMPLE

Writing the Method

A method must be written so that an investigation is repeatable by another person.

​

In order for results from an investigation to be reliable an investigation must be able to be repeated exactly the same way following the method. The results gained from each repeat must show the same pattern each time for the conclusion to be valid (or if not an explanation or fault in following the method given).

​

Your method must be repeatable by another person and include:

>independent (variable changed) and dependent (variable measured) variables that are clearly stated with units given.

>All variables listed that must be controlled (kept the same) AND how they are controlled

>Techniques used to increase accuracy (closer to actual value) and reliability (consistently the same when repeated)

​

Collecting Data

Data that is collected from an investigation can be analysed easier if placed into a clearly labelled and laid out data table.

The table must have:

A heading linked to the aim/hypothesis

Labelled quantities, units and symbols

Values (often numerical) of data collected

Data tables can also contain processed data such as results from multiple trials that have been averaged to give a more reliable value.

Reliability - Errors may occur in measurements may be reduced by taking the average of a number of readings

When collecting and measuring data in investigations, such as that for calculating speed, errors can occur. This may be due to the measuring instrument and the way it is used. Data can also be recorded incorrectly.

Repeating the investigation a number of times and averaging out the measurements can help reduce random errors. This value is called the Average.

To calculate the average/mean add the numbers together and divide the total by the amount of numbers:

​

Average = sum of numbers ÷ amount of numbers

Distance walked in 1 minute

Trial 1

Trial 2

Trial 3

Distance (m)

113 121 119

Average = (113 + 121 + 119 ) ÷ 3

= 117.7 m

Measuring in Science

Note: Weight is the result of force (gravity) acting on mass and is measured in Newton’s using a spring balance. Weight and Mass are often confused.

Converting measurements

Quantities are often measured in different scales depending upon what is most appropriate for the original size. In Science (and Mathematics) we use common prefixes to indicate the scale used.

We sometimes want to convert scales from one to another to compare data or to place the measurements into equations.

Prefix Scale

Kilo = 1000

Centi = 1/100^th

Milli = 1/1000^th

​

​

So 1 kilometre = 1000 metres

1 metre contains 100 centimetres

1 metre contains 1000 millimetres

To convert from grams to kilograms divide by 1000

(or metres to kilometres and millilitres to litres)

To convert from kilograms to grams multiply by 1000

(or kilometres to metres and litres to millilitres)

When a line graph is used to analyse data from a fair test the dependent variable (variable measured) must be placed on the y axis and the independent variable (variable changed) must be on the x axis.

​

A line of best fit is used to generate a straight line – this shows the trend and allows a gradient to be calculated.

Do not join the points

Drawing a line Graph

A line of best fit gives the smallest distance from all plotted points to the line

Independent variable verses Dependent variable graph

Calculating Gradient

Height ball dropped (cm) - independent

Height ball bounces (cm) - dependant

A line graph can be used to

calculate gradient. The co-ordinates

of a straight line in the graph are taken (for example from A to B) by projecting back to the x and y axis.

​

To calculate the value for x find the difference between t1 and t2 by subtracting the smallest value from the largest value. This will be your ∆x.

Repeat on the y axis. This will be your ∆y.

​

Gradient = rise = ∆y

run ∆x

​

The relationship of the variables is stated as a mathematical equation

Y = gradient x X

for example:

Height ball bounces(cm) = gradient x height ball dropped(cm)

∆y

∆x

12

Writing a conclusion based on the gradient

A gradient of a line will be positive when the rise of the variable changed causes the rise of the variable measured.

A gradient of a line will be negative when the rise of the variable changed caused the fall of the variable measured.

​

You must include either statement (positive or negative) in your conclusion based on your results

Writing a conclusion

A conclusion looks for patterns in collected data from an investigation.

Both the variable that is changed (independent) and the variable that is measured (dependant) must be included in the conclusion statement.

The data is used as evidence in the conclusion.

The conclusion can also be used to answer the original aim

EXAMPLE

Discussion

This part of an investigation covers what you did to increase reliability with repeats, and discussion how you kept all other variables controlled. Accuracy is discussed along with the techniques used to ensure accuracy such as reducing parallax errors and anything else to make sure your data was collected without error, such as correcting for zero error.

Areas of the investigation that could have been improved (and were modified to improve them) are discussed as well as known unavoidable errors are made.

Unexpected random results and outliers in the data can be explained, and the method used to discard them from averages.

Science ideas that could explain the results and conclusion are discussed here. Any relevant equations (including the mathematical relationship equation from the graph) can be included and further explained.

Any differences between your results and expected results based on known Science ideas can be discussed.

The discussion is an in-depth report on your investigation.

Discussion - use diagrams!

Using a diagram is a great way to show your understanding of the scientific principle:

Reliability and Accuracy in the Discussion

Reliability means that that any results produced in a scientific investigation must be more than a one-off finding and be repeatable (i.e. do 3 or more trials) and the averaging repeats

Other scientists must be able to perform exactly the same investigation using the same method and generate the same results.

​

Accuracy is the extent to which a investigation measures what it is supposed to measure. Techniques to improve: zero error, parallax error reduction, scale selection, reduce reaction time

Reliable

Not Accurate

Low Reliability

Low Accuracy

Not Reliable

Not Accurate

Both Reliable

And Accurate

17

Accuracy - Scale selection

Accuracy - Parallax

The direct line of sight when making a reading from a ruler needs to be used to avoid parallax error. Measurements made by viewing the position of a ruler (or meter) relative to something to be measured are subject to parallax error if the ruler is some distance away from the object of measurement and not viewed from directly on.

To avoid parallax error read the ruler straight on and level, as well as holding the ruler as close to the object as possible.

Accuracy - reducing errors

Whenever we measure something the measurement is never exact, it is an estimate of the value of a physical quantity.

Errors (not mistakes) can be caused by limitations to the accuracy of measurement.

There are two kinds of error:

Systematic

Caused by faulty equipment or experimental design, often affect all results to the same extent.

e.g. Friction causes an object not to accelerate as quickly as expected or a ruler may be incorrectly used.

constant attention to detail is needed to avoid systematic errors

Random

These result from the limits of the accuracy of all measuring devices. They can be reduced but can never be eliminated.

e.g. Reaction time, sensitivity of measuring apparatus or observer error/parallax error.

Accuracy - Zero error

Many rulers do not start at zero. When you are measuring with a ruler you need to account for this error, called zero error. You will need to measure the gap from the end of the ruler to the start of measuring (the zero line) and deduct that amount away from each measurement.

This normally occurs when you are measuring a height from the floor upwards.

Zero error