Study Notes on Split Plot Design

Split Plot Design in Experimental Research

Overview of the Split Plot Design

  • A split plot design is a type of experimental design used to evaluate two or more factors where only one factor is randomized.
  • In this case, we have three treatments applied to different groups of animals.

Treatments and Animals

  • Treatment A:
    • Animals involved: 1, 2, 3, 4, 5
  • Treatment B:
    • Animals involved: 6, 7, 8, 9, 10
  • Time Factor:
    • Different time points measured include 0, 60, 120, 180, and 240 minutes.

Data Collection

  • Each animal was challenged and blood samples were collected every 60 minutes.
  • The focus was on measuring LH (Luteinizing Hormone) concentrations in the blood.
  • Five samples per treatment were collected, allowing for a detailed analysis of variance across time points.

Differences from Complete Randomized Design

  • In a complete randomized design, each treatment would receive random allocation without consideration for blocking factors (e.g., time).
  • The split plot design allows for a more intricate investigation by measuring responses at multiple time points.

Calculating Sums and Squares

  • Participants were asked to compute various statistics including the sums and sums of squares for the treatments and time points.
    • Sum of X: Refers to the sum of LH concentration values at each time point.
    • Sum of X Squared: Necessary for later analysis typically involves the computation of the variance.
Example Calculations Performed by Participants

  • Animal Group Treatment A:

    • Sum of X for various time points (0, 60, 120…) calculated by participants.

  • Statistics for individuals for Treatment A:

    • Animal 1 to 5: Various sums computed, leading to values like 1.75 for total sum at specific time points.

  • Animal Group Treatment B:

    • Participants also calculated sums for animals 6 to 10 corresponding to treatment B across the same time points.
Detailed Calculations
  • The sums and sums of squares needed were specified for calculations, some of which were performed by the participants, such as:
    • Sum of X squared for Animal 1, 2, 3, etc.
    • Sum of X for each specified time point was computed accordingly.

Statistical Analysis

  • After obtaining sums and sums of squares, the discussion transitioned to analysis of variance (ANOVA).
  • Degrees of Freedom: 15 (number of animals multiplied by number of samples minus 1).
  • Total Sum of Squares:
    • Sum of X: 17240.02
    • Sum of X Squared: 113929.52
  • The computations also relied on established formulas applied to the collected data to quantify variance within and between treatment groups.

Usage of SAS in Analysis

  • The instructor emphasized the importance of software like SAS in analysis, primarily for ease and accuracy compared to manual calculations, especially when dealing with complex data involving multiple variables.
  • It was mentioned that outdated methods (manual calculations) posed significant challenges in analysis.

Key Takeaways

  • Understanding split plot design allows researchers to maximize insights from limited resources by assessing more than one factor and their interactions simultaneously.
  • Calculating sums and variances in a structured way is critical for interpreting data correctly in research methods applied to animal science.
  • The course also emphasizes statistical analysis software and its relevance for efficient research methodology in agricultural sciences.