1. INTRODUCTION

This report investigates various characteristics of the individuals in the bodyfat dataset, with the central research question being how do factors like age, weight, height, and circumference measurements relate to body fat percentage in individuals?

This research question is relevant because understanding the relationships between physical characteristics (age, weight, height, and circumferences) and body fat percentage can provide valuable insights for health professionals, fitness experts, and individuals seeking to manage or understand their body composition. Given the rise in obesity-related health issues and the importance of body fat in assessing overall health, exploring these relationships is essential for creating targeted health interventions and personal fitness plans.

The dataset consists of 100 entries with 10 variables measuring body fat percentage, age, weight, height, and several body circumferences (neck, chest, abdomen, etc.). The data is publicly available from Lock5Stat and is structured in a way that allows for various types of statistical analysis, including descriptive statistics and visualizations.

#data collection

bodyfat = read.csv("https://lock5stat.com/datasets3e/BodyFat.csv")
head(bodyfat)
##   Bodyfat Age Weight Height Neck Chest Abdomen Ankle Biceps Wrist
## 1    32.3  41 247.25  73.50 42.1 117.0   115.6  26.3   37.3  19.7
## 2    22.5  31 177.25  71.50 36.2 101.1    92.4  24.6   30.1  18.2
## 3    22.0  42 156.25  69.00 35.5  97.8    86.0  24.0   31.2  17.4
## 4    12.3  23 154.25  67.75 36.2  93.1    85.2  21.9   32.0  17.1
## 5    20.5  46 177.00  70.00 37.2  99.7    95.6  22.5   29.1  17.7
## 6    22.6  54 198.00  72.00 39.9 107.6   100.0  22.0   35.9  18.9

The data used in this analysis comes from the https://lock5stat.com/datasets3e/BodyFat.txt, which is publicly available on [https://lock5stat.com/datapage3e.html]. The dataset includes 100 individuals and contains the following variables:

Bodyfat: Body fat percentage (float) Age: Age of the individual (integer) Weight: Weight in pounds (float) Height: Height in inches (float) Neck: Neck circumference in cm (float) Chest: Chest circumference in cm (float) Abdomen: Abdomen circumference in cm (float) Ankle: Ankle circumference in cm (float) Biceps: Biceps circumference in cm (float) Wrist: Wrist circumference in cm (float)

The dataset is organized with each row representing an individual’s measurements across the listed variables. The sample size of 100 entries is appropriate for statistical analysis, allowing for meaningful insights into trends and relationships between variables. The data is structured consistently, and the variables are clearly defined, for the analysis of body fat percentage,age, height, weight, and circumferences.

#Question1: 10 question by me

  1. What is the average (mean) bodyfat percentage across all individuals in the dataset? Method: Mean

  2. Is there a correlation between age and bodyfat percentage? Method: Correlation

  3. How does the distribution of weight look across the sample? Method: Histogram

  4. What is the median height of the individuals in the dataset? Method: Median

  5. Is there a relationship between chest circumference and abdomen circumference? Method: Correlation and Scatterplot

  6. How does the bodyfat percentage vary across different age groups? Method: Boxplot (with age grouped into categories)

  7. What is the standard deviation of wrist circumference in the sample? Method: Standard Deviation

  8. How does the average weight compare between individuals above and below the median age? Method: Mean and Barplot

  9. Is there a correlation between height and weight in the dataset? Method: Correlation

  10. What is the distribution of neck circumferences in the sample? Method: Histogram or Boxplot

#Question2. 10 questions again generated

  1. Descriptive Statistics:
    • What is the average body fat percentage of the individuals in the dataset? (Mean, median)
  2. Age Distribution:
    • What is the age distribution of the individuals? (Histogram)
  3. Weight Variability:
    • What is the variability (range, standard deviation) in the weight of the individuals? (Variance, standard deviation)
  4. Height vs. Body Fat:
    • Is there a correlation between height and body fat percentage? (Correlation, scatterplot)
  5. Chest Circumference Analysis:
    • What is the spread of chest circumference measurements? (Boxplot)
  6. Abdomen vs. Weight:
    • Does abdomen circumference show a strong linear relationship with weight? (Correlation, scatterplot)
  7. Neck Size Categories:
    • How are individuals distributed across different neck circumference ranges? (Barplot or histogram)
  8. Body Fat Variance by Age Group:
    • Does the variance in body fat percentage differ between younger (<30) and older (≥30) individuals? (Variance, boxplot)
  9. Biceps Size Distribution:
    • What is the distribution of biceps circumference measurements? (Histogram or boxplot)
  10. Wrist Size Analysis:
    • How does wrist circumference compare across the dataset? (Mean, median, standard deviation)

#Question 3

Average Body Fat Percentage:

What is the average (mean) body fat percentage across all individuals in the dataset? Method: Mean, Median Age Distribution:

What is the age distribution of the individuals? Method: Histogram Weight Variability:

What is the range and standard deviation of weight in the dataset? Method: Range, Standard Deviation Height and Weight Relationship:

Is there a correlation between height and weight in the dataset? Method: Correlation, Scatterplot Chest Circumference Spread:

What is the spread of chest circumference measurements? Method: Boxplot Abdomen Circumference and Weight:

Does abdomen circumference show a strong linear relationship with weight? Method: Correlation, Scatterplot Body Fat Variance by Age Group:

Does the variance in body fat percentage differ between younger (<30) and older (≥30) individuals? Method: Boxplot, Variance Wrist Size Analysis:

How does wrist circumference compare across the dataset? Method: Mean, Median, Standard Deviation Biceps Circumference Distribution:

What is the distribution of biceps circumference measurements? Method: Histogram Height Distribution:

What is the overall distribution of height in the sample, including any outliers? Method: Boxplot

2.METHODOLOGY

The analysis was performed on a dataset consisting of 100 individuals and 10 variables, including body fat percentage, age, weight, height, and several body circumferences. The primary methods employed include:

#do the data cleaning

# Checking for missing values
sum(is.na(bodyfat))
## [1] 0
sum(is.na(bodyfat$Bodyfat))  # Missing values in Bodyfat column
## [1] 0
sum(is.na(bodyfat$Age))      # Missing values in Age column
## [1] 0
sum(is.na(bodyfat$Weight))   # Missing values in Weight column
## [1] 0
mean_bodyfat <- mean(bodyfat$Bodyfat, na.rm = TRUE)
sd_weight <- sd(bodyfat$Weight, na.rm = TRUE)
bodyfat_clean <- na.omit(bodyfat)  # Removes rows with 

bodyfat$Bodyfat[is.na(bodyfat$Bodyfat)] <- mean(bodyfat$Bodyfat, na.rm = TRUE)
bodyfat$Weight[is.na(bodyfat$Weight)] <- mean(bodyfat$Weight, na.rm = TRUE)


correlation <- cor(bodyfat$Bodyfat, bodyfat$Age, use = "complete.obs")

##RESULTS

4.1. Body Fat Percentage Method: Mean, Median

The mean body fat percentage is calculated to give an overall sense of the central tendency of the data. The median body fat percentage provides a measure that is less sensitive to extreme values.

mean_bodyfat = mean(bodyfat$Bodyfat, na.rm = TRUE)
median_bodyfat = median(bodyfat$Bodyfat, na.rm = TRUE)

cat("The mean body fat percentage is ", mean_bodyfat, ".\n", sep = "")
## The mean body fat percentage is 18.601.
cat("The median body fat percentage is ", median_bodyfat, ".\n", sep = "")
## The median body fat percentage is 18.95.

Mean Body Fat Percentage: The mean body fat percentage is approximately 18.60%, which is typical for the population in this dataset. Median Body Fat Percentage: The median is 18.95%, suggesting a slight skew in the dataset, as the mean is slightly lower than the median.

4.2. Age and Body Fat Percentage Correlation Method: Correlation

To explore the relationship between age and body fat percentage, we calculated the correlation coefficient. A positive correlation indicates that older individuals tend to have higher body fat percentages.

cor_age_bodyfat = cor(bodyfat$Age, bodyfat$Bodyfat, use = "complete.obs")
cat("The correlation between age and body fat percentage is ", cor_age_bodyfat, ".\n", sep = "")
## The correlation between age and body fat percentage is 0.2557976.

Correlation: The correlation coefficient is 0.25, indicating a weak positive correlation between age and body fat percentage. Older individuals tend to have slightly higher body fat percentages.

4.3. Distribution of Weight Method: Histogram

A histogram is used to visualize the distribution of weight in the dataset, helping to understand the spread and shape of the weight data.

hist(bodyfat$Weight, main = "Weight Distribution", xlab = "Weight (lbs)", col = "lightblue")

The weight data is left-skewed, with most individuals in the lower weight range, but a few individuals have significantly higher weights.

4.4. Median Height Method: Median

The median height is calculated to determine the central value of height in the sample.

median_height = median(bodyfat$Height, na.rm = TRUE)
cat("The median height is ", median_height, " inches.\n", sep = "")
## The median height is 70 inches.

The median height is 70 inches, indicating that most individuals in the dataset are of average height.

4.5. Relationship Between Chest and Abdomen Circumference Method: Correlation and Scatterplot

The relationship between chest and abdomen circumference is explored with a correlation coefficient and visualized through a scatterplot.

cor_chest_abdomen = cor(bodyfat$Chest, bodyfat$Abdomen, use = "complete.obs")
cat("The correlation between chest and abdomen circumference is ", cor_chest_abdomen, ".\n", sep = "")
## The correlation between chest and abdomen circumference is 0.9227279.
plot(bodyfat$Abdomen ~ bodyfat$Chest, xlab = "Chest Circumference (cm)", ylab = "Abdomen Circumference (cm)", col = "red")

Positive Correlation: The overall trend suggests a positive correlation between the two variables. This means that as chest circumference increases, abdomen circumference also tends to increase.

Moderate Strength: The correlation appears to be of moderate strength. The points are not tightly clustered around a straight line, indicating some variability.

4.6. Body Fat Percentage by Age Group Method: Boxplot (with age grouped into categories)

Body fat percentage is compared across different age groups using a boxplot, which shows the distribution and any potential outliers.

age_groups = cut(bodyfat$Age, breaks = c(20, 30, 40, 50, 60, 70), include.lowest = TRUE, labels = c("<30", "30-39", "40-49", "50-59", "60+"))
boxplot(Bodyfat ~ age_groups, data = bodyfat, xlab = "Age Group", ylab = "Body Fat Percentage")

We observe a general trend of increasing body fat percentage as age advances. The median line within each box (representing the 50th percentile) shows a gradual upward shift from younger to older age groups.

Distribution within Age Groups:

Younger Age Groups (<30, 30-39): These groups exhibit lower body fat percentages. The boxes are relatively shorter, indicating less variability in body fat within these age groups. Middle Age Groups (40-49, 50-59): These groups show a noticeable increase in body fat percentage compared to younger groups. The boxes are taller, suggesting a wider range of body fat values within these age groups. Older Age Group (60+): This group has the highest body fat percentage on average. The box is relatively tall, suggesting a significant spread in body fat values among individuals in this age group.

The whiskers in the boxplot extend to show the range of data excluding outliers. The dots beyond the whiskers represent potential outliers, indicating individuals with exceptionally high or low body fat percentages for their age group.

4.7. Standard Deviation of Wrist Circumference Method: Standard Deviation

The standard deviation of wrist circumference is calculated to assess how much variability there is in this measurement across the sample.

sd_wrist = sd(bodyfat$Wrist, na.rm = TRUE)
cat("The standard deviation of wrist circumference is ", sd_wrist, " cm.\n", sep = "")
## The standard deviation of wrist circumference is 0.9993225 cm.

The standard deviation is 0.9993 cm, indicating low variability in wrist circumference across the dataset.

4.8. Average Weight by Age Group Method: Mean and Barplot

The mean weight is compared between individuals above and below the median age, and the results are visualized with a barplot.

median_age = median(bodyfat$Age, na.rm = TRUE)
age_group = ifelse(bodyfat$Age > median_age, "Above Median", "Below Median")
avg_weight = tapply(bodyfat$Weight, age_group, mean, na.rm = TRUE)

barplot(avg_weight, main = "Average Weight by Age Group", xlab = "Age Group", ylab = "Average Weight (lbs)", col = c("blue", "green"),  ylim = c(0, 200))

median_age = median(bodyfat$Age, na.rm = TRUE)

age_group = ifelse(bodyfat$Age > median_age, "Above Median", "Below Median")

avg_weight = tapply(bodyfat$Weight, age_group, mean, na.rm = TRUE)

# Print the results
cat("Average weight for individuals above the median age is ", avg_weight["Above Median"], " lbs.\n", sep = "")
## Average weight for individuals above the median age is 174.4565 lbs.
cat("Average weight for individuals below the median age is ", avg_weight["Below Median"], " lbs.\n", sep = "")
## Average weight for individuals below the median age is 180.0028 lbs.

Average Weight: The average weight for individuals above the median age is 174.45 lbs, while it is 180.00 lbs for those below the median age, reflecting the tendency of younger individuals to weigh more.

4.9. Correlation Between Height and Weight Method: Correlation

The correlation between height and weight is calculated to understand how strongly these two variables are related.

cor_height_weight = cor(bodyfat$Height, bodyfat$Weight, use = "complete.obs")
cat("The correlation between height and weight is ", cor_height_weight, ".\n", sep = "")
## The correlation between height and weight is 0.5684328.

Correlation: The correlation coefficient is 0.56, indicating a moderate positive relationship between height and weight.

4.10. Distribution of Neck Circumference Method: Histogram or Boxplot

The distribution of neck circumference is analyzed using a histogram.

hist(bodyfat$Neck, main = "Neck Circumference Distribution", xlab = "Neck Circumference (cm)", col = "lightgreen")

The highest frequency of neck circumferences is around 35-39 cm. The lowest frequency is in the 32-35 cm range, but it gradually decreases beyond 39 cm, until 44 cm.

Analysis

The analysis revealed several meaningful insights related to the research question. The weak correlation between age and body fat percentage suggests that while age plays a role in body fat accumulation, it is not the only determining factor. Other factors, such as lifestyle choices, exercise, diet, and genetics, likely contribute to body fat variation.

The left-skewed distribution of weight suggests that the sample tends to have a lower weight range, with only a few individuals being on the higher end of the scale. This distribution may point to the sample’s overall leaner body composition, but it also highlights the variability in body weight.

The positive correlation between chest and abdomen circumferences is consistent with the hypothesis that larger body sizes tend to have larger measurements in these areas. This could indicate a general pattern of body fat distribution, where individuals with larger torsos tend to accumulate more fat in the abdomen area.

The increase in body fat percentage across different age groups supports the hypothesis that aging contributes to body fat accumulation. This finding is important for health professionals when recommending fitness and health strategies for different age groups, as older adults typically experience a natural increase in body fat percentage.

The differences in average weight between individuals below and above the median age align with expectations that younger individuals may have higher weight due to higher muscle mass and physical activity levels. These findings contribute to the understanding that body composition and weight management strategies should differ for younger and older populations.

Finally, the distribution of neck circumference is relatively standard, with most individuals falling into the 35-39 cm range. This could reflect a more typical range of body proportions, suggesting that the sample does not have extreme variations in neck circumference.

Conclusion

Based on the analysis, several conclusions can be drawn:

Age is weakly related to body fat percentage, but other factors likely play a significant role. Although body fat does increase with age in this dataset, the relationship is not strong enough to imply that age alone is the main factor affecting body fat. Lifestyle and genetic factors also play crucial roles.

Weight tends to be lower for older individuals, possibly due to reduced muscle mass and lower physical activity levels with age. Younger individuals in the dataset, on average, weigh more, which could be attributed to higher muscle mass and activity levels.

Body fat percentage increases with age, particularly in older age groups. This pattern supports health recommendations to manage body fat as individuals age, with an emphasis on maintaining muscle mass and healthy body composition.

Body circumferences (chest, abdomen, and neck) provide useful insights into body fat distribution. The correlations observed suggest that these measurements can be used to predict body fat in certain contexts, particularly when combined with other variables like weight and height.

In conclusion, the research highlights important patterns in how age, weight, height, and body circumference relate to body fat percentage. These insights can guide future research and inform health and fitness strategies, particularly for different age groups,

##Citations and References Sources:

Lock5Stat. (n.d.). BodyFat Dataset. Retrieved November 16, 2024, from https://lock5stat.com/datasets3e/BodyFat.txt

Textbook References:

STAT353 notes: https://rpubs.com/scsustat/STAT353