MMI-L2-Moodle

Lecture Overview

• Course Title: MVLS: Life Sciences Level 2

• Topic: Contemporary Issues in Biology 2 Block 2

• Lecturers: Dr. Stevie A Bain (she/her), Dr. Paul Capewell (he/him)

Intended Learning Outcomes

• Difference Between Samples and Populations: Understanding the distinction between individual samples and entire populations.

• Sampling Error: Definition and explanation of how sampling error can occur.

• Data Classification: Identify and present data as continuous or categorical appropriately.

• Statistical Hypotheses: Formulate null hypotheses (H0) and alternative hypotheses (H1) for various experiments.

• Chi-squared Test: Overview of the chi-squared test, including calculations (observed counts, expected counts, degrees of freedom, p-values). Note: R code not required for memorization.

Data Presentation

Class Height Data

• Visualization: Histogram for MMI1$Height with various heights marked.

Histograms

• Hist() Command Modifiers:

  • Number of bins: breaks

  • X-axis label: xlab

  • Y-axis label: ylab

  • Bar color: col

  • Main title: main

  • X-axis limit: xlim

  • Y-axis limit: ylim

Samples and Populations

• Definition of Sample: A sample is a very small subset of a total population.

• Population: All members constituting a defined group.

• Representativeness: Questions of how well a sample represents the population and potential sampling biases.

Investigation Approach

• Random Subsets: Conduct random sampling to perform statistical analyses (mean, median, range).

• Impact on Results: Explore whether random sampling influences outcomes.

Sampling Error

• Definition: Random variation in data arising from sampling only part of the population.

Factors Influencing Human Height

• Various influences include:

  • Genetic factors (inheritance, parental height)

  • Environmental conditions (nutrition, lifestyle)

  • Societal factors (socioeconomic status, beauty standards)

  • Biological factors (hormone levels, medical conditions)

  • Geographical location and climate influences

Biological Sex as a Modifier of Height

• Importance of collecting data on biological sex alongside height data.

• Participation in data collection is voluntary and anonymous.

Data Analysis Steps

• Clean data using Excel.

• Use R Studio for visualization and statistical analysis.

• Acknowledge differences between continuous and categorical data in presentation and analysis.

Expected Outcomes

• Discuss inferences regarding expected sex ratios based on genetic determination.

• Comparison of expected versus observed graphs will be assessed.

Importance of Statistical Testing

• Mere observation of data is insufficient; statistical testing validates findings.

• Consider if deviations from expectations result from sampling error or other factors.

Hypotheses in Statistics

• Statistical Hypothesis: An assumption about population characteristics or variable relationships subject to testing.

• Null Hypothesis (H0):

  • Default assumption that there’s no relationship or effect.

  • Categorical outcomes assumed equally likely.

• Alternative Hypothesis (H1):

  • Suggests a difference exists; categorical outcomes are not equally likely.

Sex Ratio Contextual Hypotheses

• Design hypotheses to evaluate sex ratios using:

  • H0: Male to female ratio = 1:1

  • H1: Male to female ratio ≠ 1:1

  • Further variations can be hypothesized (more females or males).

Chi-squared Test

• Application: Used for categorical data to compare observed versus expected frequencies under H0.

• Assumptions:

  • Categories are mutually exclusive.

  • Observations are independent.

  • Expected values mostly > 5.

• Formula:

  • ( \chi^2 = \sum \frac{(d^2)}{e} ) where ( d ) = observed - expected.

Contingency Table and Degrees of Freedom

• Construct a contingency table to display observed counts.

• Calculate degrees of freedom using: (rows - 1) x (columns - 1).

Calculating p-values

• Example p-value chart provided for significance levels based on the chi-squared statistic and degrees of freedom.

Using R for Analysis

• Utilization of R for expedited statistical analysis and p-value computations.

Upcoming Topics

• Tomorrow's Lecture: Focus on interpreting p-values and assess the role of genetic sex as a modifier of human height.