Edexcel S3 2017 June — Question 3 10 marks

Exam BoardEdexcel
ModuleS3 (Statistics 3)
Year2017
SessionJune
Marks10
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicHypothesis test of Spearman’s rank correlation coefficien
TypeHypothesis test for positive correlation
DifficultyStandard +0.3 This is a straightforward application of Spearman's rank correlation coefficient with clear data already in rank form. Part (a) requires calculating rs using the standard formula with d² values, part (b) is a routine one-tailed hypothesis test comparing to critical values from tables, and part (c) asks for a simple interpretation. The question involves no conceptual challenges—just methodical calculation and standard test procedure, making it slightly easier than average for S3 level.
Spec5.08e Spearman rank correlation5.08f Hypothesis test: Spearman rank

  1. A junior judge is being trained by a senior judge to learn how to assess ice skaters. After the training, the judges each assess 6 ice skaters \(A , B , C , D , E\) and \(F\). They each list them in order of preference with the best ice skater first. The results are shown in the table below.
Rank123456
Senior Judge\(A\)\(B\)\(D\)\(C\)\(F\)\(E\)
Junior Judge\(B\)\(D\)\(A\)\(F\)\(C\)\(E\)
  1. Calculate Spearman's rank correlation coefficient for these data.
  2. Test, at the \(5 \%\) level of significance, whether or not there is evidence of a positive correlation between the rankings of the junior judge and the senior judge. State your hypotheses clearly.
  3. Comment on the effectiveness of the training delivered by the senior judge.

Question 3:
Part (a):
AnswerMarks Guidance
Senior JudgeJunior Judge \(d\)
\(A\)1 3
\(B\)2 1
\(C\)4 5
\(D\)3 2
\(E\)6 6
\(F\)5 4
8
M1 for attempt to rank judges' lists (at least 4 correct for each judge)M1A1 A1 for correct rankings for both (may be reversed); can be implied by correct \(d^2\) or \(r_s\); table could be ordered in terms of Senior Judge
\(\sum d^2 = 8\) or 62A1 8 or 62 or correct \(d^2\) row
\(r_s = 1 - \frac{6 \times 8}{6 \times 35} = \frac{27}{35} = 0.771\)dM1A1 M1 for use of correct formula, follow through their \(\sum d^2\) (dependent on 1st M1); if answer is not correct, a correct expression is required; A1 exact fraction or awrt \((\pm)0.771\)
Part (b):
AnswerMarks Guidance
\(H_0: \rho = 0\)B1 Both hypotheses in terms of \(\rho\) or \(\rho_s\); hypotheses just in words e.g. "no correlation" score B0
\(H_1: \rho > 0\)
Critical value \(\pm 0.8286\)B1 Accept \(\pm 0.8857\) if 2-tailed \(H_1\)
\((0.771 < 0.8286)\) so insufficient evidence to reject \(H_0\)M1 Follow through from their \(r_s\) and their c.v. if \(
There is insufficient evidence to suggest a positive correlation between the judgesA1ft A correct contextualised comment that includes "judges"
Part (c):
AnswerMarks Guidance
For positive correlation c.v. is \(0.8286 > 0.771\)
Training of junior judge was ineffectiveB1ft Follow through from their cv and \(r_s\)
## Question 3:

### Part (a):

| | Senior Judge | Junior Judge | $d$ | $d^2$ |
|---|---|---|---|---|
| $A$ | 1 | 3 | -2 | 4 |
| $B$ | 2 | 1 | 1 | 1 |
| $C$ | 4 | 5 | -1 | 1 |
| $D$ | 3 | 2 | 1 | 1 |
| $E$ | 6 | 6 | 0 | 0 |
| $F$ | 5 | 4 | 1 | 1 |
| | | | | 8 |

M1 for attempt to rank judges' lists (at least 4 correct for each judge) | M1A1 | A1 for correct rankings for both (may be reversed); can be implied by correct $d^2$ or $r_s$; table could be ordered in terms of Senior Judge

$\sum d^2 = 8$ or 62 | A1 | 8 or 62 or correct $d^2$ row

$r_s = 1 - \frac{6 \times 8}{6 \times 35} = \frac{27}{35} = 0.771$ | dM1A1 | M1 for use of correct formula, follow through their $\sum d^2$ (dependent on 1st M1); if answer is not correct, a correct expression is required; A1 exact fraction or awrt $(\pm)0.771$

### Part (b):
$H_0: \rho = 0$ | B1 | Both hypotheses in terms of $\rho$ or $\rho_s$; hypotheses just in words e.g. "no correlation" score B0
$H_1: \rho > 0$ | |

Critical value $\pm 0.8286$ | B1 | Accept $\pm 0.8857$ if 2-tailed $H_1$

$(0.771 < 0.8286)$ so insufficient evidence to reject $H_0$ | M1 | Follow through from their $r_s$ and their c.v. if $|cv| < 1$ and $|r_s| < 1$

There is insufficient evidence to suggest a **positive** correlation between the judges | A1ft | A correct contextualised comment that includes "judges"

### Part (c):
For positive correlation c.v. is $0.8286 > 0.771$ | |

Training of junior judge was ineffective | B1ft | Follow through from their cv and $r_s$

---
\begin{enumerate}
  \item A junior judge is being trained by a senior judge to learn how to assess ice skaters. After the training, the judges each assess 6 ice skaters $A , B , C , D , E$ and $F$. They each list them in order of preference with the best ice skater first. The results are shown in the table below.
\end{enumerate}

\begin{center}
\begin{tabular}{ | l | c | c | c | c | c | c | }
\hline
Rank & 1 & 2 & 3 & 4 & 5 & 6 \\
\hline
Senior Judge & $A$ & $B$ & $D$ & $C$ & $F$ & $E$ \\
\hline
Junior Judge & $B$ & $D$ & $A$ & $F$ & $C$ & $E$ \\
\hline
\end{tabular}
\end{center}

(a) Calculate Spearman's rank correlation coefficient for these data.\\
(b) Test, at the $5 \%$ level of significance, whether or not there is evidence of a positive correlation between the rankings of the junior judge and the senior judge. State your hypotheses clearly.\\
(c) Comment on the effectiveness of the training delivered by the senior judge.

\hfill \mbox{\textit{Edexcel S3 2017 Q3 [10]}}