Podcast
Questions and Answers
Which of these can be a factor that shows internal homogeneity? (Select all that apply)
Which of these can be a factor that shows internal homogeneity? (Select all that apply)
The ______ is a procedure that is designed to support, refute, or validate a hypothesis.
The ______ is a procedure that is designed to support, refute, or validate a hypothesis.
experiment
Experimental factors are conditions we impose on the experimental units in order to determine the cause and effect of explanatory variables on the variable of interest?
Experimental factors are conditions we impose on the experimental units in order to determine the cause and effect of explanatory variables on the variable of interest?
True
Blocking factors can be thought of as conditions that are confounding characteristics that the experimental units are assigned with.
Blocking factors can be thought of as conditions that are confounding characteristics that the experimental units are assigned with.
Signup and view all the answers
The 'control variable' needs to be controlled so that we can properly observe the effect of the independent variable?
The 'control variable' needs to be controlled so that we can properly observe the effect of the independent variable?
Signup and view all the answers
Which of the following is NOT a type of experimental design? (Select all that apply)
Which of the following is NOT a type of experimental design? (Select all that apply)
Signup and view all the answers
The 'independent variable' in an experiment is the variable of interest, and the focus of your hypothesis?
The 'independent variable' in an experiment is the variable of interest, and the focus of your hypothesis?
Signup and view all the answers
It is NOT important to control for any other variables in experimental design?
It is NOT important to control for any other variables in experimental design?
Signup and view all the answers
Which of the following is NOT a Hill's criteria for Causality?
Which of the following is NOT a Hill's criteria for Causality?
Signup and view all the answers
Which of these is NOT a way to establish causality? (Select all that apply)
Which of these is NOT a way to establish causality? (Select all that apply)
Signup and view all the answers
What criteria can be used to determine whether an association between two variables is likely to be causal?
What criteria can be used to determine whether an association between two variables is likely to be causal?
Signup and view all the answers
Study Notes
ENV330 Lecture 3: More on Basic Experimental Designs
- Key elements of experimental designs include whether the predictor variable is manipulated (e.g., by applying a treatment) or observational (something observed only, not manipulated), whether experimental units are randomly assigned to treatments, and what data is collected about responses and explanatory variables (e.g., categoricals, quantitives)
- Five key components of experimental designs are: controlled conditions, experimental controls, randomization, replication and blocking
- Controlled experiments with categorical explanatory variables (factors) can use completely randomized designs (1 factor), randomized block designs (1 factor), pairs designs (1 factor), and 2+ factor designs
- Subsets and spatial arrays, randomized blocks, Latin square, split plots, covariates and repeated measures are examples of experimental designs
- A completely randomized design example involves randomly assigning mice to two different diets (diet A and diet B). Comparing the results on means and variance is an aspect of the study, but if the sex of the mice may affect the results, then a different study design may be needed.
- A randomized block design accounts for blocking variables, an example involves considering mouse sex as a blocking factor when assessing diet effects on mice
- Matched pairs are a type of blocked design (BACI) and use the same experimental units in two treatments (diet A and diet B) to control for individual variability. This helps prevent unmeasured factors from influencing the results
- BACI (before, after, control, impact) designs are a type of blocked experiment that includes a control area to monitor the data
- Experimental factors are the factors that are imposed on experimental units to measure some effect on the dependent variable
- Blocking factors affect experimental units but aren't the things of interest, instead are aspects affecting the study that are controlled
- The 'factor' is the explanatory variable experimented with, or imposed on the experimental units. The 'level' is the specific version or category of that factor
- Examples of factors are categories like water treatment or type of inoculation; levels include quantities for something like water treatment in quantities of being well or water stressed, or amount or type of inoculation
- Multiple factors affect multiple variables, and can have multiple types of outcomes, additive or interactive
- Spatial arrays use blocking to consider spatial heterogeneity when experimental units are spatially distributed
- An example relates to studying the effect of fertilizer on seed production in two rice varieties
- Completely randomized block design (RCBD) ensures variation in blocks rather than within blocks
- RCBD should be relatively homogenous, but also large enough to ensure treatment is well separated in space
- Spatial arrays are useful when environmental gradients are found across a spatial area; if a variable varies significantly across a spatial gradient, but is internally homogenous then blocking is useful
- Latin square designs are a type of spatial array with every row and column having all the treatments
- A split-plot experiment involves two factors, one 'easy' factors (e.g., irrigation) and one 'hard' factor (e.g., fertilizer), applied to plots and sub-plots in the field
- Covariates need to be measured instead of controlled for, a continuous variable that might affect the result (not neatly grouping into neat, homogenous blocks) such as soil conditions • Covariates help statistically control for individual variability, example being leaf area vs. leaf size to accurately quantify effect on the plant from factors like tar spots
- Repeated measures designs include measuring the same experimental entity over time, accounting for the time-to-time dependencies (autocorrelation)
- Data structures are often hierarchical (e.g., fish, trees, streams, birds) and require sophisticated approaches
- Dependent variables depend on other variables
- Independent variables are things that cause or affect the outcome of some process
- Covariates are variables that you want to control or manage to prevent any unmeasured things from having an effect
- Independent variables are factors (and covariates) controlling or preventing any effect on outcomes from other unmeasured factors: independent factors, control factors, or blocking factors
- A complete set of factors and covariates should be included
- Different factors can have fixed or random effects depending on the hypothesis you wish to test
- Categorical and levels are factors, and can vary across settings
- Summary elements: Hypothesis testing is used to establish causal relationships among independent and dependent variables. Understanding the contribution of different factors and controlling for confounding variables are important
- Additional discussion: The distinction between observational (sampling) and manipulative (experimentation) studies in environmental science
- Key takeaway elements from lecture notes and other supplemental material presented
- Summary for this lecture and additional discussion
- Introduction to field exercises and activities
Studying That Suits You
Use AI to generate personalized quizzes and flashcards to suit your learning preferences.
Related Documents
Description
Explore the key elements of experimental designs in this quiz covering randomization, replication, and various design types such as completely randomized and blocked designs. Understand how manipulating predictor variables affects the outcome and learn about the data collection methods involved. This quiz will help solidify your understanding of controlled experimental conditions and their applications.
