Problem 5 - An Introduction to Psychological Statistics

Question

Finish filling out the following ANOVA tables: $$ \text { a. } K=4 $$ $$ \begin{array}{lllll} \text { Source } & S S & d f & M S & F \\ \hline \text { Between } & 87.40 & & \\ \text { Within } & & & \\ \text { Total } & 199.22 & 33 & & \\ & & & & \\ \hline \end{array} $$ b. \(N=14\) \begin{tabular}{lllll} Source & \(S S\) & \(d f\) & \(M S\) & \(F\) \\ \hline Between & & 2 & \(14.10\) & \\ Within & & & & \\ & & & \\ Total & \(64.65\) & & & \end{tabular}

Step-by-Step Solution

Answer

Problem (a): F = 7.81 Problem (b): F = 4.26

Step 1: Fill Out 'Between' Degrees of Freedom for Problem (a)

The degrees of freedom for the 'Between' group in ANOVA is given by \( df_{between} = K - 1 \), where \( K \) is the number of groups. Here, \( K = 4 \), so \( df_{between} = 4 - 1 = 3 \).

Step 2: Fill Out 'Within' and 'Total' Degrees of Freedom for Problem (a)

The total degrees of freedom is \( df_{total} = 33 \). We can use the formula for total degrees of freedom \( df_{total} = df_{between} + df_{within} \). So, \( df_{within} = df_{total} - df_{between} = 33 - 3 = 30 \).

Step 3: Compute Mean Square (MS) and F for Problem (a)

The Mean Square for 'Between' is calculated as \( MS_{between} = \frac{SS_{between}}{df_{between}} = \frac{87.40}{3} \approx 29.13 \). To find the Mean Square for 'Within', we need the \( SS_{within} \), which is \( 199.22 - 87.40 = 111.82 \). Therefore, \( MS_{within} = \frac{111.82}{30} = 3.73 \). Now, \( F = \frac{MS_{between}}{MS_{within}} = \frac{29.13}{3.73} \approx 7.81 \).

Step 4: Fill Out 'Total' Degrees of Freedom for Problem (b)

The problem states the total sample size \( N = 14 \) and there are 3 groups (implied from \( df_{between} = 2 \), since \( df_{between} = K - 1 \)). Thus, the total degrees of freedom are \( df_{total} = N - 1 = 14 - 1 = 13 \).

Step 5: Compute 'Within' and 'Between' Sum of Squares for Problem (b)

Using \( MS = \frac{SS}{df} \), for 'between' we have \( SS_{between} = MS_{between} \times df_{between} = 14.10 \times 2 = 28.20 \). Total Sum of Squares \( SS_{total} = 64.65 \), so \( SS_{within} = SS_{total} - SS_{between} = 64.65 - 28.20 = 36.45 \).

Step 6: Compute Mean Square (MS) and F for Problem (b)

Now that we know \( SS_{within} = 36.45 \) and \( df_{within} = df_{total} - df_{between} = 13 - 2 = 11 \), we can calculate \( MS_{within} = \frac{36.45}{11} = 3.31 \). Finally, \( F = \frac{MS_{between}}{MS_{within}} = \frac{14.10}{3.31} \approx 4.26 \).

Key Concepts

Degrees of Freedom

When analyzing any dataset using ANOVA (Analysis of Variance), understanding the concept of degrees of freedom (df) is crucial. It helps in defining the flexibility we have in spreading out our data points. Moreover, it is essential for calculating other key statistics like the mean square and the F-statistic.

In ANOVA, there's usually more than one type of degrees of freedom involved:

• Between Groups: It quantifies the variance due to the differences between the groups. It's calculated as \( df_{\text{between}} = K - 1 \), where \( K \) is the number of groups.
• Within Groups: This tells us about the variability among subjects within each group. It's calculated using the formula \( df_{\text{within}} = df_{\text{total}} - df_{\text{between}} \).
• Total: It reflects the overall sample size, resulting in \( df_{\text{total}} = N - 1 \), where \( N \) is the total number of observations.

Each type of degrees of freedom plays a distinct role in the analysis and their correct computation is necessary for accurate outcomes.

Mean Square

The Mean Square, abbreviated as MS, is a crucial part of the ANOVA analysis, which essentially averages the types of variability within your data. Calculating MS helps in understanding how varied the data points are within and between groups.

Here's how it's performed:

• Mean Square Between (MSB): Represents the variance between the group means and is calculated using the formula \( MS_{\text{between}} = \frac{SS_{\text{between}}}{df_{\text{between}}} \). This value provides an average of the differences across group means.
• Mean Square Within (MSW): Averages the variance within each group. It is computed as \( MS_{\text{within}} = \frac{SS_{\text{within}}}{df_{\text{within}}} \), giving an idea of variability inside each group.

Both these mean squares are vital as they contribute to the computation of the F-statistic, thus helping determine if there are significant differences between the group means.

F-statistic

The F-statistic is a major output of ANOVA tests, which provides insight into whether any of the group means differ significantly. It uses both the Mean Square Between and Mean Square Within for its computation.

Here's the formula:

• \( F = \frac{MS_{\text{between}}}{MS_{\text{within}}} \)

The F-statistic is essentially a ratio of the variability between the group means to the variability within the groups. A higher value suggests that the group means are significantly different from each other.

It is important to compare this calculated F-value against a critical F-value from the F-distribution table. If the calculated F is higher, it typically indicates significant differences among group means.

The F-statistic thus acts as a marker for hypothesis testing, allowing us to infer whether the variation between groups can be attributed to random chance or actual differences.

Sum of Squares

In ANOVA, the Sum of Squares (SS) is a measure of data variability and is split into different sources. By partitioning this variability, we can better understand how much variation is due to differences between group means and how much is due to differences within the groups themselves.

The main components are:

• Sum of Squares Between (SSB): Is attributed to variance due to differences among group means. You calculate it indirectly through \( SS_{\text{between}} = MS_{\text{between}} \times df_{\text{between}} \) when Mean Square Between is provided.
• Sum of Squares Within (SSW): Measures variability due to differences within each group. It's calculated similarly using \( SS_{\text{within}} = MS_{\text{within}} \times df_{\text{within}} \).
• Total Sum of Squares (SST): Sum of all variancies, given as \( SS_{\text{total}} = SSB + SSW \), representing overall variability in the dataset.

Understanding these components sheds light on the overall variance and assists in assessing the significance of the mean differences detected through the ANOVA.