OCR MEI Further Statistics Major 2022 June — Question 10 13 marks

Exam BoardOCR MEI
ModuleFurther Statistics Major (Further Statistics Major)
Year2022
SessionJune
Marks13
PaperDownload PDF ↗
Mark schemeDownload PDF ↗
TopicChi-squared test of independence
TypeExpected frequencies partially provided
DifficultyStandard +0.3 This is a straightforward chi-squared test question requiring calculation of expected frequencies using the standard formula (row total × column total / grand total) and computing one contribution to the test statistic using (O-E)²/E. These are routine procedures taught in any statistics course with no conceptual difficulty or novel insight required.
Spec5.06a Chi-squared: contingency tables

10 A scientist is researching dietary fat intake and cholesterol level. A random sample of 60 people is selected and their dietary fat intakes and cholesterol levels are measured. Dietary fat intakes are classified as low, medium and high, and cholesterol levels are classified as normal and high. The scientist decides to carry out a chi-squared test to investigate whether there is any association between dietary fat intake and cholesterol level. Tables \(\mathbf { 1 0 . 1 }\) and \(\mathbf { 1 0 . 2 }\) show the data and some of the expected frequencies for the test. \begin{table}[h]
\multirow{2}{*}{}Dietary fat intake
LowMediumHighTotal
\multirow{2}{*}{Cholesterol level}Normal918532
High3131228
Total12311760
\captionsetup{labelformat=empty} \caption{Table 10.1}
\end{table} \begin{table}[h]
Expected frequencyDietary fat intake
\cline { 3 - 5 }LowMediumHigh
\multirow{2}{*}{
Cholesterol
level
}
Normal9.0667
\cline { 2 - 5 }High7.9333
\captionsetup{labelformat=empty} \caption{Table 10.2}
\end{table}
  1. Complete the table of expected frequencies in the Printed Answer Booklet.
  2. Determine the contribution to the chi-squared test statistic for people with normal cholesterol level and high dietary fat intake, giving your answer to \(\mathbf { 4 }\) decimal places. The contributions to the chi-squared test statistic for the remaining categories are shown in Table 10.3. \begin{table}[h]
    Dietary fat intake
    \cline { 2 - 5 }LowMediumHigh
    \multirow{2}{*}{
    Cholesterol
    level
    }
    Normal1.05630.1301
    \cline { 2 - 5 }High1.20710.14872.0846
    \captionsetup{labelformat=empty} \caption{Table 10.3} \end{table}
  3. In this question you must show detailed reasoning. Carry out the test at the 5\% significance level.
  4. For each level of dietary fat intake, give a brief interpretation of what the data suggest about the level of cholesterol.

  5. Question 10:
    AnswerMarks Guidance
    10(a) 32×12 28×12
    Low-normal= =6.4, Low-high= =5.6
    60 60
    32×31
    Medium-normal= =16.5333
    60
    28×31
    Medium-high= =14.4667
    AnswerMarks
    60B1
    B1
    AnswerMarks
    [2]3.4
    1.1B1 for any one correct
    B1 for the other 3 (by subtraction)
    AnswerMarks Guidance
    10(b) (5−9.0667)2
    9.0667
    AnswerMarks
    = 1.8240M1
    A1
    AnswerMarks
    [2]1.1
    1.1(O−E)2
    For
    E
    AnswerMarks Guidance
    10(c) DR
    H : no association between dietary fat intake and
    0
    cholesterol level
    H : some association between dietary fat intake and
    1
    cholesterol level
    Test statistic = 6.4508
    Degrees of freedom = 2
    Critical value = 5.991
    6.4508 > 5.991 so reject H
    0
    There is sufficient evidence to suggest that there is some
    association between dietary fat intake and cholesterol
    AnswerMarks
    level.B1
    B1
    B1
    B1
    M1
    A1
    AnswerMarks
    [6]2.5
    1.1
    3.3
    1.1
    2.2b
    AnswerMarks
    3.5aFor both Allow independent/not independent. Do NOT allow
    relationship in place of association
    BC Accept awrt 6.45
    Can be implied by correct critical value
    Allow χ2 (6.4508)=0.9603
    2
    With the above allow comparison with 95% so reject for H M1
    0
    Must have correct test statistic and critical value
    If hypotheses wrong way around allow MAX B0B1B1B1M0A0
    AnswerMarks Guidance
    10(d) For low dietary fat, the contribution of 1.0563 shows
    that more people than expected have normal
    cholesterol level whereas the contribution of 1.2071
    shows that fewer than expected have high cholesterol
    level.
    For medium dietary fat, the numbers are as expected.
    For high dietary fat, the contribution of 1.8240 shows
    that fewer people than expected have normal
    cholesterol level whereas the contribution of 2.0846
    shows that more than expected have high cholesterol
    AnswerMarks
    levelB3,2,1
    ,0
    AnswerMarks
    [3]2.2b
    3.5a
    AnswerMarks
    3.5aB3 for 5 correct contribution comments
    B2 for 4 correct contribution comments
    B1 for two correct comments
    Allow slightly fewer with higher and slightly more with normal
    Question 10:
    10 | (a) | 32×12 28×12
    Low-normal= =6.4, Low-high= =5.6
    60 60
    32×31
    Medium-normal= =16.5333
    60
    28×31
    Medium-high= =14.4667
    60 | B1
    B1
    [2] | 3.4
    1.1 | B1 for any one correct
    B1 for the other 3 (by subtraction)
    10 | (b) | (5−9.0667)2
    9.0667
    = 1.8240 | M1
    A1
    [2] | 1.1
    1.1 | (O−E)2
    For
    E
    10 | (c) | DR
    H : no association between dietary fat intake and
    0
    cholesterol level
    H : some association between dietary fat intake and
    1
    cholesterol level
    Test statistic = 6.4508
    Degrees of freedom = 2
    Critical value = 5.991
    6.4508 > 5.991 so reject H
    0
    There is sufficient evidence to suggest that there is some
    association between dietary fat intake and cholesterol
    level. | B1
    B1
    B1
    B1
    M1
    A1
    [6] | 2.5
    1.1
    3.3
    1.1
    2.2b
    3.5a | For both Allow independent/not independent. Do NOT allow
    relationship in place of association
    BC Accept awrt 6.45
    Can be implied by correct critical value
    Allow χ2 (6.4508)=0.9603
    2
    With the above allow comparison with 95% so reject for H M1
    0
    Must have correct test statistic and critical value
    If hypotheses wrong way around allow MAX B0B1B1B1M0A0
    10 | (d) | For low dietary fat, the contribution of 1.0563 shows
    that more people than expected have normal
    cholesterol level whereas the contribution of 1.2071
    shows that fewer than expected have high cholesterol
    level.
    For medium dietary fat, the numbers are as expected.
    For high dietary fat, the contribution of 1.8240 shows
    that fewer people than expected have normal
    cholesterol level whereas the contribution of 2.0846
    shows that more than expected have high cholesterol
    level | B3,2,1
    ,0
    [3] | 2.2b
    3.5a
    3.5a | B3 for 5 correct contribution comments
    B2 for 4 correct contribution comments
    B1 for two correct comments
    Allow slightly fewer with higher and slightly more with normal
    10 A scientist is researching dietary fat intake and cholesterol level. A random sample of 60 people is selected and their dietary fat intakes and cholesterol levels are measured. Dietary fat intakes are classified as low, medium and high, and cholesterol levels are classified as normal and high.
    
    The scientist decides to carry out a chi-squared test to investigate whether there is any association between dietary fat intake and cholesterol level. Tables $\mathbf { 1 0 . 1 }$ and $\mathbf { 1 0 . 2 }$ show the data and some of the expected frequencies for the test.
    
    \begin{table}[h]
    \begin{center}
    \begin{tabular}{|l|l|l|l|l|l|}
    \hline
    \multicolumn{2}{|c|}{\multirow{2}{*}{}} & \multicolumn{3}{|c|}{Dietary fat intake} &  \\
    \hline
     &  & Low & Medium & High & Total \\
    \hline
    \multirow{2}{*}{Cholesterol level} & Normal & 9 & 18 & 5 & 32 \\
    \hline
     & High & 3 & 13 & 12 & 28 \\
    \hline
     & Total & 12 & 31 & 17 & 60 \\
    \hline
    \end{tabular}
    \captionsetup{labelformat=empty}
    \caption{Table 10.1}
    \end{center}
    \end{table}
    
    \begin{table}[h]
    \begin{center}
    \begin{tabular}{ | c | l | c | c | c | }
    \hline
    \multicolumn{2}{|c|}{Expected frequency} & \multicolumn{3}{c|}{Dietary fat intake} \\
    \cline { 3 - 5 }
     & Low & Medium & High &  \\
    \hline
    \multirow{2}{*}{\begin{tabular}{ c }
    Cholesterol \\
    level \\
    \end{tabular}} & Normal &  &  & 9.0667 \\
    \cline { 2 - 5 }
     & High &  &  & 7.9333 \\
    \hline
    \end{tabular}
    \captionsetup{labelformat=empty}
    \caption{Table 10.2}
    \end{center}
    \end{table}
    \begin{enumerate}[label=(\alph*)]
    \item Complete the table of expected frequencies in the Printed Answer Booklet.
    \item Determine the contribution to the chi-squared test statistic for people with normal cholesterol level and high dietary fat intake, giving your answer to $\mathbf { 4 }$ decimal places.
    
    The contributions to the chi-squared test statistic for the remaining categories are shown in Table 10.3.
    
    \begin{table}[h]
    \begin{center}
    \begin{tabular}{ | c | l | c | c | c | }
    \hline
    \multicolumn{2}{|c|}{\begin{tabular}{ c }
    Contribution to the \\
    test statistic \\
    \end{tabular}} & \multicolumn{3}{|c|}{Dietary fat intake} \\
    \cline { 2 - 5 }
     & Low & Medium & High &  \\
    \hline
    \multirow{2}{*}{\begin{tabular}{ c }
    Cholesterol \\
    level \\
    \end{tabular}} & Normal & 1.0563 & 0.1301 &  \\
    \cline { 2 - 5 }
     & High & 1.2071 & 0.1487 & 2.0846 \\
    \hline
    \end{tabular}
    \captionsetup{labelformat=empty}
    \caption{Table 10.3}
    \end{center}
    \end{table}
    \item In this question you must show detailed reasoning.
    
    Carry out the test at the 5\% significance level.
    \item For each level of dietary fat intake, give a brief interpretation of what the data suggest about the level of cholesterol.
    \end{enumerate}
    
    \hfill \mbox{\textit{OCR MEI Further Statistics Major 2022 Q10 [13]}}