Connect sequences and growth

Distinguish additive and multiplicative change, discrete indexing and continuous extensions.

What to practice

  • Distinguish additive and multiplicative change, discrete indexing and continuous extensions.

Unit 2 begins with a simple question: when the input increases by a fixed step, does the output change by adding the same amount or by multiplying by the same factor? Additive change produces arithmetic sequences and linear functions. Multiplicative change produces geometric sequences and exponential functions. This chapter treats sequences as functions on whole-number inputs (topic 2.1) and then connects them to the linear and exponential functions that extend them to every real input (2.2). Chapter 8 builds on this foundation with the full exponential family.

Quick prerequisite check. What is the slope of the line through (2,5)(2,5) and (6,17)(6,17), and what is 3⋅243\cdot 2^4? If you answered 3 and 48, you are ready. The first is a rate of change; the second is repeated multiplication.

Sequences are functions on whole-number inputs

A sequence is a function whose inputs are whole-number indices. The notation ana_n means the term with index nn. Two choices must be made explicit: the first index (often 0 or 1) and what each term represents. A sequence's graph is a set of separate points, one per index, because no term exists between index 3 and index 4.

TypeConstant featureExplicit rule from a known term aka_kStep-by-step rule
arithmeticdifference ddan=ak+d(n−k)a_n=a_k+d(n-k)an=an−1+da_n=a_{n-1}+d
geometricratio rr, with r≠0r\neq 0gn=gk⋅rn−kg_n=g_k\cdot r^{n-k}gn=r⋅gn−1g_n=r\cdot g_{n-1}

The explicit rules work from any known term, not only the first one, which saves time when a problem gives two terms in the middle of a sequence. Each step of the index adds dd or multiplies by rr, so moving from index kk to index nn takes n−kn-k steps.

Worked example 7.1: find a sequence from two terms

Arithmetic. An arithmetic sequence has a3=14a_3=14 and a7=30a_7=30. Four steps separate the two terms, so 4d=30−14=164d=30-14=16 and d=4d=4. Then an=14+4(n−3)=4n+2a_n=14+4(n-3)=4n+2. Check: 4(7)+2=304(7)+2=30.

Geometric. A geometric sequence has g2=12g_2=12 and g5=96g_5=96. Three steps separate the terms, so r3=9612=8r^3=\frac{96}{12}=8 and r=2r=2. Then gn=12⋅2n−2g_n=12\cdot 2^{n-2}, which gives g1=6g_1=6. The common error is to divide the change by the number of steps, which works only for differences, not ratios.

Worked example 7.2: choose the starting index on purpose

A ball is dropped from a height of 10 m, and each bounce reaches 60% of the previous height. Two correct rules describe the same heights:

  • Index from 0, with h0=10h_0=10 the drop height: hn=10(0.6)nh_n=10(0.6)^n is the height of bounce nn.
  • Index from 1, with b1=6b_1=6 the first bounce: bn=6(0.6)n−1b_n=6(0.6)^{n-1}.

Both give the height of the fourth bounce as 10(0.6)4=6(0.6)3=1.29610(0.6)^4=6(0.6)^3=1.296 m. Mixing the two, for example writing 10(0.6)n−110(0.6)^{n-1} for bounce nn, shifts every answer by one bounce. State the index meaning in words before writing the rule.

A negative ratio needs care. The sequence gn=5(−2)ng_n=5(-2)^n gives 5,−10,20,−40,…5,-10,20,-40,\ldots: the signs alternate and the absolute values double. It is neither increasing nor decreasing. A ratio between −1-1 and 0 also alternates, but the terms shrink toward 0.

Squares for the arithmetic sequence rising by 3 at each index and open circles for the geometric sequence multiplying by 1.5, for indices 1 to 8. The circles start lower but first rise above the squares at index 7.

Figure 7.1. Equal index steps produce a constant difference for an=4+3(n−1)a_n=4+3(n-1) and a constant ratio for gn=2(1.5)n−1g_n=2(1.5)^{n-1}. The points are not joined, because a sequence has no terms between indices.

Errors with sequences

  • Starting the exponent at the wrong index, which shifts every term by one step.
  • Finding rr by dividing the change by the number of steps; ratios require a root.
  • Calling a sequence with a negative ratio decreasing.
  • Drawing a sequence as a continuous curve without saying that this is a modeling choice.

Try it 7.1. A sequence has first term 81 and fourth term 24. Find the sixth term if the sequence is geometric, and again if it is arithmetic.

Check your work. Geometric: r3=2481=827r^3=\frac{24}{81}=\frac{8}{27}, so r=23r=\frac{2}{3} and the sixth term is 81(23)5=323≈10.6781\left(\frac{2}{3}\right)^5=\frac{32}{3}\approx 10.67. Arithmetic: 3d=24−81=−573d=24-81=-57, so d=−19d=-19 and the sixth term is 81+5(−19)=−1481+5(-19)=-14. The same two terms lead to very different predictions, which is why a model's type must be justified.

Equal input steps reveal additive or multiplicative change

The patterns from sequences carry over to functions on all real inputs. For a linear function f(x)=b+mxf(x)=b+mx, increasing the input by any fixed amount Δ\Delta changes the output by the same amount, mΔm\Delta. For an exponential function g(x)=a⋅bxg(x)=a\cdot b^x with a≠0a\neq 0, increasing the input by Δ\Delta multiplies the output by the same factor, bΔb^{\Delta}. Sampled at equally spaced inputs, a linear function produces an arithmetic sequence and an exponential function produces a geometric sequence.

In this sense, linear functions are continuous versions of arithmetic sequences, and exponential functions are continuous versions of geometric sequences. Whether a continuous extension is appropriate depends on the context. Water draining from a tank changes at every instant, so a function of time on an interval is reasonable. Members counted once each week exist only at whole weeks, so the sequence is the honest model, and the curve between the points is a convenience for seeing the trend.

Read the parameters in context

In f(x)=b+mxf(x)=b+mx, bb is the output when the input is 0, and mm is the change in output for each 1-unit increase in input, measured in output units per input unit. In g(x)=a⋅bxg(x)=a\cdot b^x, aa is the output when the input is 0, and bb is the factor that multiplies the output for each 1-unit increase. A base b>1b>1 means growth of (b−1)⋅100(b-1)\cdot 100 percent per unit; a base between 0 and 1 means decay of (1−b)⋅100(1-b)\cdot 100 percent per unit. If a tank holds V(t)=500−12tV(t)=500-12t liters after tt minutes, it loses 12 liters every minute. If a town's population is P(t)=2000(0.97)tP(t)=2000(0.97)^t after tt years, it shrinks by 3 percent each year, so it loses fewer people each year as it gets smaller.

Worked example 7.3: classify from an equally spaced table

xx036912
f(x)f(x)713192531
g(x)g(x)46913.520.25

The inputs step by 3. The outputs of ff change by 6 each time, so ff is linear with slope 63=2\frac{6}{3}=2 and f(x)=7+2xf(x)=7+2x. The outputs of gg do not change by a constant amount, but each is 1.5 times the previous one, so gg is exponential with g(x)=4(1.5)x/3g(x)=4(1.5)^{x/3}. The factor 1.5 belongs to each 3-unit step; the factor for a 1-unit step is the cube root of 1.5, about 1.145.

Worked example 7.4: unequal spacing hides a linear pattern

xx1247
h(x)h(x)10131928

The output changes are 3, 6 and 9, which are not constant. The input changes are 1, 2 and 3, however, so each average rate of change is 31=62=93=3\frac{3}{1}=\frac{6}{2}=\frac{9}{3}=3. The function is consistent with a linear model, h(x)=7+3xh(x)=7+3x. Compare differences or ratios only across equal input steps.

Worked example 7.5: two points, two models

Find a linear function and an exponential function through (1,6)(1,6) and (3,54)(3,54), and compare them at x=5x=5.

Linear. The slope is 54−63−1=24\frac{54-6}{3-1}=24, so f(x)=6+24(x−1)f(x)=6+24(x-1) and f(5)=102f(5)=102.

Exponential. Two input steps of 1 multiply the output by 9, so b2=9b^2=9 and, because an exponential base is positive, b=3b=3. Then a⋅3=6a\cdot 3=6 gives a=2a=2, so g(x)=2⋅3xg(x)=2\cdot 3^x and g(5)=486g(5)=486.

Two points always determine one model of each type, so two points alone never decide which type is correct. Use a third observation or a reason from the context.

Worked example 7.6: when does multiplicative growth overtake additive growth?

In a constructed practice scenario, club A has An=3000+500(n−1)A_n=3000+500(n-1) members in week nn, and club B has Bn=800(1.25)n−1B_n=800(1.25)^{n-1}. In week 11, A11=8000A_{11}=8000 and B11≈7451B_{11}\approx 7451. In week 12, A12=8500A_{12}=8500 and B12≈9313B_{12}\approx 9313. Club B first has more members in week 12, as Figure 7.2 shows. An increasing exponential pattern eventually exceeds any linear pattern, even one with a large head start, because its additions keep growing while the linear additions stay at 500.

Squares for club A rising by 500 each week and open circles for club B growing by 25 percent each week, for weeks 1 to 14, with dotted curves through each set. The circles stay below the squares through week 11 and are above them from week 12 on.

Figure 7.2. Constructed practice data. The markers are the weekly values; the dotted curves are continuous extensions included only to show the trend.

Errors with linear and exponential change

  • Applying a per-step factor to a different step length, such as using 1.5 per unit when 1.5 belongs to each 3-unit step.
  • Reading unequal input steps as equal ones.
  • Treating a percent increase as a fixed amount added each period.
  • Reporting a non-whole answer, such as week 11.4, when the sequence exists only at whole indices.

Try it 7.2. For x=0,2,4,6x=0,2,4,6, a function has outputs 40, 30, 22.5 and 16.875. Decide whether it is linear or exponential, write a rule, and predict the output at x=8x=8.

Check your work. The differences are −10-10, −7.5-7.5 and −5.625-5.625, so the function is not linear. Each output is 0.75 times the previous one over each 2-unit step, so it is exponential: f(x)=40(0.75)x/2f(x)=40(0.75)^{x/2}. Then f(8)=16.875(0.75)≈12.66f(8)=16.875(0.75)\approx 12.66.

Common errors to repair

ErrorRepair
Dividing a change by the step count to find a ratioWorked example 7.1
Writing the exponent from the wrong starting indexWorked example 7.2
Applying a per-step factor to a different step lengthWorked example 7.3
Comparing differences across unequal input stepsWorked example 7.4
Reporting a non-whole week for a sequenceWorked example 7.6

Review and retrieval

Answer from memory first, then check against the chapter.

  1. Write explicit rules for an arithmetic and a geometric sequence from a known term aka_k.
  2. Explain why a linear function sampled at equal input steps produces an arithmetic sequence.
  3. Describe how to test a table with unequal input spacing.
  4. Give one context where a continuous extension is reasonable and one where only the sequence is.

Retry Try it 7.1 at your next session without notes, then rebuild Figure 7.1 from its two rules a few days later. Chapter 8 uses these ideas to build exponential models from data and percent change.

Practice

Twelve multiple-choice questions (Questions 117–128) are printed after this chapter. Questions 117–126 are no-calculator questions; work them by hand. Questions 127 and 128 require a graphing calculator. Keys, complete reasoning for every choice and repair directions are under Chapter 7: answers and explanations in the book's answers section.

Use this chapter companion

Review the lesson, attempt the original practice, then check the explanation. Your printed study guide remains complete without the website.

Independence

Skill by Skill Prep is independent. SAT® and AP® are trademarks registered by the College Board, which is not affiliated with, and does not endorse, this website or book. PSAT/NMSQT® is a registered trademark of the College Board and the National Merit Scholarship Corporation, which are not affiliated with, and do not endorse, this site. Bluebook™ is a trademark of the College Board. ACT® is a trademark of ACT Education Corp., which is not affiliated with and does not endorse this website or book. Desmos is a trademark of Desmos Studio PBC. Practice results describe these questions, not official scores or admission predictions.

Last updated .