EST 523 First Exam Topic List
Explain the following statistical terminology: bias, precision, accuracy, consistency, parameter and statistic
Be able to select a simple random sample, stratified random sample, systematic random sample and estimate the population mean or total.
State the advantages a stratified design has over a SRS,
State the limitations of unreplicated systematic sample and how variance estimation is usually performed.
Describe how confidence intervals are constructed, the assumptions underlying them, etc.
Describe and distinguish between measures of frequency, density, percent cover and biomass (Higgins et al. in Bookhout 1994 reading)
Describe the different measurement methods for the above variables. Emphasis will be placed on qualities such as bias and precision, sample size, characteristics that must be considered in choosing a measurement method (e.g., quadrat shape and size, plant distribution). (Krebs 64-71)
Use Wiegerts method to select optimal quadrat size, understand its components. (Krebs 68,70)
Compute Goodness of Fit tests for spatial distribution using quadrat counts. (Krebs)
Explain the connection between spatial distribution of plants and statistical distributions used to examine the physical distribution. (Krebs 72-81)
Compute and interpret index of dispersion and use it in formal hypothesis testing. (Krebs 76)
Describe the advantages, disadvantages and assumptions underlying plotless methods. (Bonham 148-165)
Compute an index and test for spatial pattern using closest individual and nearest neighbor data. (Krebs 136-140)
Explain how and why Diggles estimator(s) are more robust to nonrandom spatial patterns (Krebs 140-143)
Summarize data from PCQ method and estimate variables of interest including relative cover, relative frequency and relative density. (Handout)
Ssummarize data from line-intercept method and estimate variables of interest (Handout, Mueller-Dumbois and Ellenburg reading)
Summarize data from point-intercept method and estimate variables of interest (Handout, Mueller-Dumbois and Ellenburg reading)
Be a critical thinker.
EST 523 Second Exam Topic List
Explain what population indices are, how they are typically used, and why they are used
Describe the key assumption underlying their (index) use.
Distinguish between direct and indirect classes of indices and give examples of each.
Describe the factors that affect the proportionality of population indices to population size (e.g.,
Anderson 2001, Engeman 2003, Anderson 2003 articles).
Describe how the collection of population estimation methods (capture recapture, distance methods, removal methods, etc.) can be seen as generalizations of index methods.
Distance Sampling (Buckland et al. 1993)
Explain how Distance methods are generalizations of strip transect methods.
State the main assumptions underlying distance methodology.
Contrast the advantages and disadvantages of lines versus point transects.
Describe the design considerations involved in properly executing distance sampling.
Explain the purpose of the detectability function and the procedure for developing it.
State the advantages of defining a probability density function from the detectability function.
Describe the principle of maximum likelihood estimation.
Describe the methods of model selection that are commonly used in distance modeling.
Describe the methods for estimating variance and confidence interval construction of estimated density.
EST 523 Third Exam Topic List
Describe how the Petersen estimator adjusts an index for detectability.
Describe the assumptions underlying the Petersen estimator.
Explain the advantages of Chapmans version of Petersens estimator.
Describe how the assumptions are likely to be violated and what effect that has on the estimator.
Consider how you would design a survey to reduce the possibility of assumption violations.
Closed population models with multiple marking periods.
Recognize the parameterization of the each of the models Mo, Mt, Mb, i.e., what parameters are estimated.
What estimation method is used for the above set of models?
Explain the basis for the jackknife estimator for model Mh. i.e., what estimate is modified for bias?
Explain how model Mb eliminates the effect of trap response in its estimation process.
Given a set of data, could you use a least squares regression approach for estimating N?
Model Mbh, explain how it removes the effect of heterogeneity and trap response.
Describe the techniques and considerations that are available to assist in selecting an appropriate model.
Explain the parsimony principle.
Open population model
Jolly-Seber model estimators of parameters, explain what they are computing.
State the assumptions underlying the use of this model
What are the effects of assumption violations?
Explain how the marked population size is estimated (e.g., what groups of animals are followed and what is measured?)
Describe the structure of the data for this design.
What are its advantages over a strictly open design approach (e.g., Jolly-Seber model).
State the assumptions underlying the method, compute an estimate of N, and recognize when the estimator fails.
Explain the connection between the removal method and model Mb and the differences in their estimation methods.
Explain how the generalized removal estimator handles heterogeneity.