| Exam Board | Edexcel |
|---|---|
| Module | S3 (Statistics 3) |
| Year | 2014 |
| Session | June |
| Marks | 12 |
| Paper | Download PDF ↗ |
| Mark scheme | Download PDF ↗ |
| Topic | Hypothesis test of Spearman’s rank correlation coefficien |
| Type | Hypothesis test for positive correlation |
| Difficulty | Standard +0.3 This is a standard S3 hypothesis testing question requiring calculation of Spearman's rank correlation coefficient and performing routine hypothesis tests. The ranking is straightforward (no ties in x), the test procedure is formulaic using critical value tables, and part (c) uses a given PMCC value. While it requires multiple steps and careful execution, it demands only direct application of learned procedures with no novel insight or complex reasoning. |
| Spec | 5.08a Pearson correlation: calculate pmcc5.08d Hypothesis test: Pearson correlation5.08e Spearman rank correlation5.08f Hypothesis test: Spearman rank |
| Man | \(\boldsymbol { A }\) | \(\boldsymbol { B }\) | \(\boldsymbol { C }\) | \(\boldsymbol { D }\) | \(\boldsymbol { E }\) | \(\boldsymbol { F }\) | \(\boldsymbol { G }\) | \(\boldsymbol { H }\) | \(\boldsymbol { I }\) | \(\boldsymbol { J }\) |
| \(x\) | 123 | 128 | 137 | 143 | 149 | 153 | 154 | 159 | 162 | 168 |
| \(w\) | 78 | 93 | 85 | 83 | 75 | 98 | 88 | 87 | 95 | 99 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer/Working | Mark | Guidance |
| Ranks for \(x\): 1,2,3,4,5,6,7,8,9,10 and \(w\): 2,7,4,3,1,9,6,5,8,10 (or reversed) | M1 | Attempt to rank both \(x\) and \(w\); at least four correct |
| \(\sum d^2 = 1+25+1+1+16+9+1+9+1+0 = 64\) | M1, A1 | Finding differences between ranks and evaluating \(\sum d^2\); \(\sum d^2 = 64\) |
| \(r_s = 1 - \frac{6(64)}{10(99)}\ ; = 0.6121212...\ \left(\frac{101}{165}\right)\) | dM1; A1 | Using \(1 - \frac{6\sum d^2}{10(99)}\) with their \(\sum d^2\); awrt 0.612. Dependent on 1st M1. If \(\sum d^2 = 266\), \(r_s = \text{awrt}\ {-0.612}\): award M1M1A1M1A1 |
| Answer | Marks | Guidance |
|---|---|---|
| Answer/Working | Mark | Guidance |
| \(H_0: \rho = 0\), \(H_1: \rho > 0\) | B1 | Both hypotheses stated correctly in terms of \(\rho\) |
| Critical value \(r_s = 0.5636\) or CR: \(r_s \geqslant 0.5636\) | B1 | |
| Since \(r_s = 0.6121...\) lies in the CR; Result is significant; Reject \(H_0\) | M1 | Correct statement relating their \(r_s\) (\( |
| Conclude there is a positive correlation between systolic blood pressure and weight | A1 | Must mention "positive correlation", "blood pressure" and "weight". Follow through their \(r_s\) with their c.v. (provided \( |
| Answer | Marks | Guidance |
|---|---|---|
| Answer/Working | Mark | Guidance |
| Critical value \(r = 0.5494\); CR: \(r \geqslant 0.5494\); and since \(r = 0.5114\) does not lie in CR; result is not significant; do not reject \(H_0\) | M1 | |
| Conclude there is no positive correlation | A1 | Context not required |
| Answer | Marks | Guidance |
|---|---|---|
| Answer/Working | Mark | Guidance |
| "As \(x\) increases, \(w\) increases" and "the relationship is non-linear"; or "There is a positive correlation" and "the relationship is non-linear"; or "Data is not (bi-variate) normal" | B1 | Any one of these or equivalent |
# Question 4:
## Part (a)
| Answer/Working | Mark | Guidance |
|---|---|---|
| Ranks for $x$: 1,2,3,4,5,6,7,8,9,10 and $w$: 2,7,4,3,1,9,6,5,8,10 (or reversed) | M1 | Attempt to rank both $x$ and $w$; at least four correct |
| $\sum d^2 = 1+25+1+1+16+9+1+9+1+0 = 64$ | M1, A1 | Finding differences between ranks and evaluating $\sum d^2$; $\sum d^2 = 64$ |
| $r_s = 1 - \frac{6(64)}{10(99)}\ ; = 0.6121212...\ \left(\frac{101}{165}\right)$ | dM1; A1 | Using $1 - \frac{6\sum d^2}{10(99)}$ with their $\sum d^2$; awrt 0.612. Dependent on 1st M1. If $\sum d^2 = 266$, $r_s = \text{awrt}\ {-0.612}$: award M1M1A1M1A1 |
## Part (b)
| Answer/Working | Mark | Guidance |
|---|---|---|
| $H_0: \rho = 0$, $H_1: \rho > 0$ | B1 | Both hypotheses stated correctly in terms of $\rho$ |
| Critical value $r_s = 0.5636$ or CR: $r_s \geqslant 0.5636$ | B1 | |
| Since $r_s = 0.6121...$ lies in the CR; Result is significant; Reject $H_0$ | M1 | Correct statement relating their $r_s$ ($|r_s|<1$) with their c.v. where $|\text{c.v.}|<1$ |
| Conclude there is a positive correlation between systolic blood pressure and weight | A1 | Must mention "positive correlation", "blood pressure" and "weight". Follow through their $r_s$ with their c.v. (provided $|\text{c.v.}|<1$). Use of "association" is A0. Two-tailed test: max B0B1M1A0 |
## Part (c)
| Answer/Working | Mark | Guidance |
|---|---|---|
| Critical value $r = 0.5494$; CR: $r \geqslant 0.5494$; and since $r = 0.5114$ does not lie in CR; result is not significant; do not reject $H_0$ | M1 | |
| Conclude there is no positive correlation | A1 | Context not required |
## Part (d)
| Answer/Working | Mark | Guidance |
|---|---|---|
| "As $x$ increases, $w$ increases" **and** "the relationship is non-linear"; or "There is a positive correlation" **and** "the relationship is non-linear"; or "Data is not (bi-variate) normal" | B1 | Any one of these or equivalent |
---
4. In a survey 10 randomly selected men had their systolic blood pressure, $x$, and weight, $w$, measured. Their results are as follows
\begin{center}
\begin{tabular}{ | c | c | c | c | c | c | c | c | c | c | c | }
\hline
Man & $\boldsymbol { A }$ & $\boldsymbol { B }$ & \multicolumn{1}{|c|}{$\boldsymbol { C }$} & \multicolumn{1}{c|}{$\boldsymbol { D }$} & \multicolumn{1}{c|}{$\boldsymbol { E }$} & $\boldsymbol { F }$ & $\boldsymbol { G }$ & $\boldsymbol { H }$ & $\boldsymbol { I }$ & $\boldsymbol { J }$ \\
\hline
$x$ & 123 & 128 & 137 & 143 & 149 & 153 & 154 & 159 & 162 & 168 \\
\hline
$w$ & 78 & 93 & 85 & 83 & 75 & 98 & 88 & 87 & 95 & 99 \\
\hline
\end{tabular}
\end{center}
\begin{enumerate}[label=(\alph*)]
\item Calculate the value of Spearman's rank correlation coefficient between $x$ and $w$.
\item Stating your hypotheses clearly, test at the $5 \%$ level of significance, whether or not there is evidence of a positive correlation between systolic blood pressure and weight.
The product moment correlation coefficient for these data is 0.5114
\item Use the value of the product moment correlation coefficient to test, at the $5 \%$ level of significance, whether or not there is evidence of a positive correlation between systolic blood pressure and weight.
\item Using your conclusions to part (b) and part (c), describe the relationship between systolic blood pressure and weight.
\end{enumerate}
\hfill \mbox{\textit{Edexcel S3 2014 Q4 [12]}}