Scaling using logarithms, linearizing data with logs + interpreting: notes and practice questions
- Logarithmic scales convert exponentially increasing values (e.g., 1, 10, 100) to a manageable linear scale (e.g., 0, 1, 2) by using logarithms.
- Semi-log graphs have one logarithmic axis and one linear axis.
- Log-log graphs have both axes on a logarithmic scale, plotting against .
- To linearise a power relationship :
- Take logarithms of both sides:
- Plot against (log-log graph) to get a straight line.
- The gradient of this line is .
- The y-intercept of this line is .
- To linearise an exponential relationship :
- Take logarithms of both sides:
- Plot against (semi-log graph) to get a straight line.
- The gradient of this line is .
- The y-intercept of this line is .
- Linearisation techniques work identically with natural logarithm or base-10 logarithm , but consistency is crucial.
- When estimating from a log-linearised graph:
- Convert the known value to its logarithm.
- Read the corresponding logarithmic value from the graph.
- Inverse the logarithm to find the actual value:
- If , then .
- If , then .
- Use GDC to plot original data and model curve to visually check fit.
- Store exact regression parameters in GDC memory to avoid compounding rounding errors in predictions.
- Always confirm the logarithm base ( or ) used in the question for correct inverse operations.
How it is examined
The discriminating idea is which transform straightens which relationship: a semi-log plot ( against ) is a straight line when , and a log-log plot ( against ) is a straight line when . Students then read the gradient and intercept back into , or . Since they are never asked to draw the graph, a question either supplies the plot or supplies the linear regression output. This is one of the places AHL 1.9 is actually needed.
- Scale very large or small numbers using logarithms.
- Linearise data using logarithms, to determine whether the data has an exponential or a power relationship, using best-fit straight lines to determine parameters.
- Interpret log-log and semi-log graphs.
In examinations, students will not be expected to draw or sketch log-log or semi-log graphs. They read and interpret them.
Linking questions
- Other contexts: growth of bacteria, or of traffic to websites and social media. Exponential graphs that show alarming absolute figures but reasonable rates of growth.
- Links to other subjects: pH semi-log curves and finding activation energy from experimental data (chemistry); exponential decay (physics); experimental work (sciences).
- TOK: does the applicability of knowledge vary across the different areas of knowledge? What would the implications be if the value of all knowledge was measured solely in terms of its applicability?
- Links to websites: Gapminder makes use of log-log graphs, www.gapminder.org.
Practice questions
8 questions · 8 mediumQuestion 1
MediumPaper 1 · calculator5 marksThe following table shows the number of days, , since a new mobile application was launched and the percentage of its initial active user base, , remaining at the beginning of that day.
| Days since launch () | 2 | 5 | 9 | 14 | 18 | 22 |
|---|---|---|---|---|---|---|
| Percentage of active users left () | 85 | 48 | 32 | 24 | 20 | 17 |
The following table shows the natural logarithm of both and on these days, rounded to 2 decimal places.
| 0.69 | 1.61 | 2.20 | 2.64 | 2.89 | 3.09 | |
|---|---|---|---|---|---|---|
| 4.44 | 3.87 | 3.47 | 3.18 | 3.00 | 2.83 |
Use the data in the second table to find the value of and the value of for the regression line, .
Assuming that the model found in part (a) remains valid, estimate the percentage of active users remaining when .
Use your GDC to perform a linear regression on the transformed data ( as the independent variable and as the dependent variable). Remember to round your answers to an appropriate number of significant figures, usually three.
Substitute into the regression equation from part (a) to find , then convert back to .
Question 2
MediumPaper 1 · calculator6 marksA biologist is studying the growth of a certain bacterial population, , over time, (in hours). She created a scatter plot with on the x-axis and on the y-axis.
The biologist noticed that the points had a strong linear correlation, so she drew a line of best fit, as shown in the diagram. The line passes through the points and .

Find an equation for in terms of .
The biologist also investigates the relationship between the population size, , and the nutrient level, . She believes that the data can be modelled by , where and are constants, and she decides to create a scatter plot to verify her belief.
State what expression the biologist should plot on each axis to verify her belief.
The scatter plot from part (b) has a linear relationship, and the biologist finds and .
Find an equation for in terms of .
First, find the equation of the line in the form , where and . Then, use the definition of logarithms to express in terms of .
Consider the form of a linear equation, . How can the given model be rearranged to fit this form?
You have an equation for in terms of from part (a) and an equation for in terms of from part (b) with given constants. Substitute one into the other and rearrange to solve for .
Question 3
MediumPaper 1 · calculator6 marksA team of engineers is studying the relationship between the input power (P, in Watts) and the output efficiency (E, as a percentage) of a new prototype device. When they plot their experimental data on a log-log graph, they observe that the data points form a perfect straight line. Given that the line passes through the points (3, 12.0411) and (6, 20.9648), find the equation of the relationship connecting P and E. Your final answer should not include logarithms.
When data forms a straight line on a log-log graph, the relationship between the original variables is a power function, . You can either linearize the equation by taking logarithms of both sides and find the slope and y-intercept, or use regression analysis on the original points.
Question 4
MediumPaper 1 · calculator7 marks(a) The growth of a bacterial colony's area, , over time, , is being studied under specific conditions. The area is measured in square micrometers () at the end of each hour for five consecutive hours. The data is recorded in Table 1.
**Table 1: Time (hours) and Area ()**
| Time (hours) | Area () |
|---|---|
| 1 | 52 |
| 2 | 140 |
| 3 | 265 |
| 4 | 398 |
| 5 | 560 |
It is believed that the growth of the colony can be modelled by an equation of the form .
Use power regression on your graphic display calculator to find the value of and the value of . Give your answers to three significant figures.
(b) The values of and can be transformed such that and . Table 2 shows data for and to three decimal places.
**Table 2: Transformed data and **
| 0.000 | 3.951 |
| 0.693 | 4.942 |
| 1.099 | 5.580 |
| 1.386 | 5.986 |
| 1.609 | 6.328 |
Find the linear regression equation of on , in the form . Give the values of and to three decimal places.
(c) Hence, show that this linear regression is equivalent to the power regression found in part (a).
Remember to input the time values into one list and the area values into another list on your GDC. Then, use the power regression function (often denoted as PwrReg or ).
Input the transformed and values into your GDC. Use the linear regression function (often denoted as LinReg or ). Be careful with rounding to three decimal places.
Recall the properties of logarithms, specifically and . You'll need to convert the linear regression equation back into the form .
Question 5
MediumPaper 2 · calculator16 marksA school administrator is tracking the spread of a rumour among students. The number of students, , who have heard the rumour is recorded hours after it began. The first results obtained are shown in the top two rows of the table below.
Complete the last row of the table, giving your answers to three decimal places.
Draw a graph of against , using appropriate scales on the axes.
Hence, state the type of model that best fits the data displayed in part (b).
The school counsellor suggests that the number of students who have heard the rumour can be modelled by .
Explain why the school counsellor is correct.
Hence, determine the values of the parameters and of the school counsellor's model, showing all your calculations. Give your answers to three significant figures.
Use your GDC to calculate the base-10 logarithm of each value. Remember to round to three decimal places.
Plot the points from the completed table in part (a). Make sure to label your axes correctly and choose a suitable scale for both and .
Observe the shape of the graph you drew in part (b). What kind of relationship does it represent?
Consider the relationship between and the linear model you identified in part (c) by taking the logarithm of both sides.
Use linear regression on the data to find the slope and y-intercept of the line of best fit. Remember that the slope corresponds to and the y-intercept corresponds to .
Question 6
MediumPaper 1 · calculator10 marksThe intensity of sound, (in W/m), from a point source varies inversely as the square of the distance, (in metres), from the source.
When m, the intensity W/m.
Find an equation for in terms of .
Find the value of when m.
Identify which two of the following graphs would form a straight line.
A. against
B. against
C. against
D. against
E. against
F. against
For each of these two graphs identified in part (b.i), find the value of the gradient of the line formed.
Recall that 'varies inversely as the square of' means for some constant . Use the given values to find .
Use the equation found in part (a.i) and substitute the new value of .
Consider the equation . How can you transform this equation or its variables to get a linear relationship of the form ?
For the graph of against , consider the form . For the graph of against , apply logarithms to the original equation and rearrange to .
Question 7
MediumPaper 1 · calculator7 marksIn a physics experiment, the relationship between the intensity of light, (in candela), and the distance from the source, (in meters), is modelled by the equation , where and are constants.
To determine and , experimental values of and are obtained. A graph of against shows a straight line passing through the points and .
Find the value of and of .
Start by linearizing the given equation using logarithms. Remember that and . The linearized form will resemble a straight line equation , where the gradient and y-intercept can be found from the given points.
Question 8
MediumPaper 1 · calculator7 marksA cup of hot tea is placed in a room with a constant ambient temperature of 22°C. The temperature of the tea, in degrees Celsius, is recorded at various times, in minutes.
It is assumed that the temperature difference, , follows an exponential decay model. To linearize the data, a graph of is plotted against . This graph is a straight line that passes through the points and .
(a) Find the equation of the straight line in the form , where and are constants.
(b) Hence,
(i) find an expression for in terms of , writing your answer in the form .
(ii) calculate the temperature of the tea, , when minutes.
To find the equation of a straight line, you first need to calculate its gradient using the two given points. Then, use one of the points and the gradient to find the y-intercept.
To convert from a logarithmic equation to an exponential one, remember that if , then . Apply the laws of exponents to separate the terms in the power.
First, use your expression from part (b)(i) to find the value of the temperature difference, , at . Then, use the definition to find the temperature .
No question on this page matches those filters. Try another difficulty or paper.
Every Scaling using logarithms, linearizing data with logs + interpreting question, marked for you
Every answer is marked mark by mark, IB-style, and the AI tutor helps when you are stuck.
Where marks are lost
- Using your own wrong value after failing a "show that." All follow through is withdrawn for the rest of that question.