In Class Questions Flashcards by Rachel Kraus

A simple random sample can be WITH or WITHOUT _____ (this will impact how to estimate the variance of the population)

replacement

How well did you know this?

Not at all

Perfectly

In the description of the scientific method from the text which form of causation (SUFFICIENT or NECESSARY) specifies that the cause “C” under consideration alone insures the occurrence of the effect “E”?

Sufficient

How well did you know this?

Not at all

Perfectly

According to the authors of the text what are two keys to a successful model?

(1) Clearly define model objectives
- predictive? mechanistic? compare scenarios?
- who will be using it?
(2) Include in the model only those features necessary to model objectives
**Remember, models are only meant to be abstractions

How well did you know this?

Not at all

Perfectly

What are two procedures for estimating a distribution theta that are discussed in Chapter 4?

(1) Method of Moments
(2) Maximum Likelihood estimators

How well did you know this?

Not at all

Perfectly

What are the 3 measures of how good an estimate is in the sense of being close to the parameter of interest?

(1) Bias
(2) Precision (sampling variance)
(3) Accuracy (combines bias and precision)

How well did you know this?

Not at all

Perfectly

Would duck eggs per nest be considered an example of a discrete or a continuously distributed response variable?

Discrete (unless you are making omelets)

How well did you know this?

Not at all

Perfectly

Why can’t the method of moments technique for estimating probability about populations of interest be used to assess reliability of estimates and perform tests of statistical significance?

Because this approach lacks information on the underlying distribution functions and thus all information comes from the sample

How well did you know this?

Not at all

Perfectly

What advantages do Maximum Likelihood Estimators gain as sample sizes increase?

Asymptotically, the estimators become less biased, variance estimates become smaller, and the distribution approaches normal in shape

How well did you know this?

Not at all

Perfectly

What are two processes for obtaining representative samples that the book emphasizes when estimating parameters?

Simple random and stratified

How well did you know this?

Not at all

Perfectly

How is a stratified random sample different from a simple random sample?
(what is an advantage of stratified?)

Stratified random sample
- sampling done in 2 stages
stage 1: choose strata of subpopulations that is assumed to be homogeneous
stage 2: draw a simple random sample from each STRATUM

ADV: estimates have greater precision if strata are selected “intelligently”

Simple random sample
- one sample

How well did you know this?

Not at all

Perfectly

Why would one want to modify the basic exponential model to allow for periodic population collapse?

Real systems have limits and thus populations crash when limits are exceeded

How well did you know this?

Not at all

Perfectly

What are two assumptions of the basic exponential growth model from the text?
- Name ALL

(1) No immigration or emigration
(2) Non-overlapping generations
(3) Constant birth and death rates
(4) No genetic variation in birth or death rates
(5) No effect of age, size, or sex
(6) No density-dependent mechanisms altering the population growth response

How well did you know this?

Not at all

Perfectly

What is the premise behind a double sampling scheme? Provide an example

An auxiliary or surrogate variable, y, is measured on a sample of n’ units
The primary variable of interest, x, is measured on a subset n of these n’ units, where n &laquo_space;n’
Values of x, are then predicted from the y, using ratio regression methods

How well did you know this?

Not at all

Perfectly

How (in words not in equations) can habitat effects be incorporated into the exponential model? (what two factors that influence the exponential model might be changed by habitat)

Birth and death rates

How well did you know this?

Not at all

Perfectly

The basic logistic model modifies the exponential model so that the per captia growth rate from “r” is equal to what value as population approaches K (carrying capacity)?

How well did you know this?

Not at all

Perfectly

Will incorporating time lags into the basic logistic model cause population size to approach carrying capacity FASTER or SLOWER?

FASTER…population size will not compensate for approaching K as quickly this results in populations overshooting K and oscillating

How well did you know this?

Not at all

Perfectly

Estimating expected lifetime remaining, individuals require which of the following: survival rates do not vary over time OR a stable age distribution?

Survival rates are constant

How well did you know this?

Not at all

Perfectly

In life table analyses, survivorship functions can describe two temporal patterns…What are they and what distinguishes these functions?

(1) Birth Flow- continuous
(2) Birth Pulse- discrete

How well did you know this?

Not at all

Perfectly

What two factors of influence does the Euler-Lotka equation make explicit when it is used to estimate lambda?

(1) Survivorship
(2) Reproduction

How well did you know this?

Not at all

Perfectly

How can the equation _____________ be written in convenient mathematical notation to account for a multi-cohort model of both survivorship and reproduction?

By using Matrix Multiplication Notation

How well did you know this?

Not at all

Perfectly

TRUE or FALSE: In an age-based model (Leslie matrix), the initial population size will change the rate of convergence to a stable age distribution but not whether such convergence occurs?

TRUE

How well did you know this?

Not at all

Perfectly

Which form of matrix projection model (Leslie/age-based or Lefkovitch/stage-based) allows transition probabilities to go in both directions?

Lefkovitch/ stage-based
-Why? Leslie/age-based: can’t get younger

How well did you know this?

Not at all

Perfectly

Lefkovitch approach (section 8.5) modifies Leslie approach by considering transition probabilities based upon ____ instead of age?

size class (or stages)

How well did you know this?

Not at all

Perfectly

Below are three potential sampling schemes for a study area with three different habitat types. Which description BEST fits the middle scheme?
*Middle is split into 3 sections with same amount of dots in each section
(a) simple random (b) systematic (c) stratified with equal allocation (d) cluster (e) double

c) stratified with equal allocation

How well did you know this?

Not at all

Perfectly

Which of the following assumptions are NOT shared by the simplest versions of both the logistic and exponential models of population growth? (a) constant birth and death rates (b) no time lags (c) no immigration/ emigration (d) no age structure to population (c) no genetic structure to population

(a) constant birth and death rates

Which of the following models is not useful for estimating survival? (a) mayfield (b) binomial (c) known fates (d) a,b, and c (e) Lincoln Peterson

(e) Lincoln Peterson

Lotka-Volterra competition assumes each population grows in a logistic fashion and the effect of competition is to do WHAT to the effective carrying capacities of both species?

lower them

The magnitude of the stable oscillations in Lotak-Volterra predator-prey system around the continuous non-trivial equilibrium point depend upon?

Initial population size

Lotak-Volterra predation equation can be modeled in discrete time OR in continuous time. Will this choice of how to model time steps influence the stability of the oscillations in the size of the predator-prey populations?

Yes (1) continuous time: will provide stable oscillations while, (2) discrete time: will create lags (as in logistic model) in response that lead to unstable oscillations

Morris and Dorak discuss TWO main goals of PVA. What are these goals and can you provide specific examples of each? *what is PVA?*

PVA = Population Viability Analysis (1) ASSESSMENTS - in which we ask how is the population doing - what is the risk of extinction? - comparing risk b/w two populations - analyzing and synthesizing monitoring data (2) MANAGEMENT - in which we ask how can we reduce risk to the population - ID key life stages or demographic processes - How large a reserve is needed - Set harvest limits - How many populations are needed

The premise of sustainable yield is that a proportion of a resource can be removed without causing long-term resource depletion. To accomplish this managers strive to maintain the per capita harvest rate equal to WHAT aspect of the population?

It's per capita growth rate

Section 11.1.3 discusses 3 models for relating harvest rates to survivorship in a population (compensatory mortality hypothesis, additive mortality hypothesis, & partial compensation hypothesis). Which of these models is assumed to operate in the absences of density-dependent mechanisms?

additive mortality hypothesis

Bessinger and McCullough discuss US federal agencies that have adopted PVA as part of their activities. Can you name one of these agencies and discuss how or why they are using PVA?

PVA = Population Viability Analysis (1) USFS - Forest Mgmt Act (1976) - mandates the maintenance of viable populations of terrestrial vertebrates - PVA 7 Forest Planning (2) USFWS - Endangered Species Act - PVA quantifying risk from projects - Providing criteria for delisting (3) Yellowstone National Park - Grizzlies - Craighead et al (1973, McCullough (1978,1986), Shaffer (1981)

Metapopulation models are concerned with predicting the fate of _____ instead of the fate of a particular population?

A population of populations (or a metapopulation)

For a basic logistic model, using either continuous or discrete time steps, what size does maximum sustainable yield maintain the population at?

K/2

If population density of the animals under investigation impacts the performance of the index in a _____ way, than indices are typically of little use?

non-linear

Metapopulation dynamics are often predicted with simulation models because analytical models become intractable. What is a limitation to the use of simulation models in these investigations?

Biological models and model parameters must often be identified in the absence of critical field data.

What should the basis be for a decision about whether to use an index or a formal estimation procedure?

Relative importance of costs vs inferential strength

Distance data allows one to estimate the probability of _____ which in turn can be used to adjust the incomplete counts for detectability when estimating densities.

detection

Soule acknowledged that doing PVA's on keystone species may be overly naïve but he argues that it may also be the most practical approach. What were two of his criteria for choosing species to perform PVA upon?

(1) Species whose activities create habitat for several other species (2) Mutualist species whose behavior enhances the fitness (e.g. reproduction, dispersal) of other species (3) predatory/ parasitic species that regulate populations of other species and whose absence would ultimately lead to a decrease in species diversity (4) Species that have spiritual, aesthetic, recreational, or economic values to humans (5) Rare or endangered species

Distance-based estimates of density assume that animals ______ are certain to be observed?

Directly on the transect line

Distance-based analysis approaches incorporate distance into their density estimates as a means of accounting for variation in ______?

Detectability

Estimators of the distribution of distances in a key function approach rely upon two components: a key function that is the starting point for estimating the detection function COMBINED WITH a _______ function for series expansion?

Series expansion OR adjustment

The idea in both least-squares and maximum likelihood approaches to catch per-unit effort estimates of population size, is to characterize ____________ at each point in time in terms of sampling effort and the size of the population under investigation.

captures

All of the least-squares approaches are based off a regression approach to estimating original population size AND they all rely on an assumption that the capture probability (p) for a time step can be estimated as a function of unit of effort required to gather that sample. Underlying this assumption is a further assumption that _____________ is small?

pi the capture probability

Change in ratio approaches provide useful estimates only when animals of different types are ___(removed or not removed)___ in proportion to their original abundance in the population.

Not removed

The classic Lincoln-Peterson approach to MRR is appropriate when sampling with _______ periods with relatively _______ intervals between them?

Two, short

Lincoln-Peterson MRR models assume populations are closed and also ______? *Hint there are two appropriate answers*

(1) No tag loss (2) Equal probability of capture

Models estimating the population size of exploited populations were described as a special case of a modified MRR approach that accounts for one of the three assumptions (time, behavior, or heterogeneity) about detectability. Which one?

behavior

Section 14.2 discusses a number of variations on K-sample capture-recapture models these include: Mh, Mtb, Mbh, Mth, and Mbh. What do these subscripted values (h, tb, bh, th, tbh) indicate about these different models?

They indicate: which of the assumptions about detectability the model is designed to account for

A failure to estimate the effective trapping area of a grid when using MRR approaches to estimate density in a closed population will cause your estimate to (over or under) estimate the actual population density.

Over

Section 14.3 talks about using program distance to estimate densities but this approach is predicated upon trapping sampling not in a grid but instead in a ______?

Web

A major limitation of estimating abundance using resighting data and telemetered animals is that recaptures of animals are observed without actually capturing the animals. A major advantage is that _________?

Resightings are cheaper and easier to obtain

How many estimators does program NOREMARK provide for mark resighting estimates of abundance?

The Sweitzer et al. example used a specific technology for resightings of the pigs in the field whose density it estimated with NOREMARK. What was that technology?

Automatic camera systems/ camera traps

Chapter 1 of MacKenzie et al. emphasizes methods for estimating occupancy based upon ________ data *Hint: binary response*

Presence/absence

MacKenzie et al.'s chapter discusses issues relating to good sampling procedures in science and frames that discussion around 3 basic questions to be addressed during the design of a sampling scheme. What are those questions?

Why? What? How?

Virtually all abundance estimation methods involve different approaches to estimation of the ______ probability followed by (or integrated with) application of that expression?

detection

Next week, we will discuss using MRR data to estimate survivorship. The models presented today are relevant to studies that follow nests through time or track the fate of animals wearing telemetry devices through time. A key distinction for these kinds of studies is that the subjects can be visited ____ with certainty of detection as either dead or alive.

repeatedly

A common way to estimate nest success is based on the binomial model (discussed in section 15.3.1), however, in most studies nests are found at various ages (since eggs laid). The corrected form of this model that accounts for age from the first observation is called the _______ method after the scientist who developed it.

Mayfield

Mortality studies of radioed animals are viewed as similar to nest survival studies in the sense that the probability of encounter is assumed to be 1. What is a critical way in which these telemetry studies differ from nest survival studies?

No natural endpoint, such as nestlings fledging

Models discussed on last Wednesday in class (e.g. Mayfield) assume constant survival among individuals & over time. The alternative approached (developed in section 15.4) estimates survival based upon time to __________.

Failure

What is the advantage to specifying survival overtime by assigning a hazard function that takes advantage of the mathematical association between S(t) and h(t)?

S(t) = survival function, the probability of survival beyond time (t) h(t) = hazard function, the instantaneous rate at which events occur at time (t), given survival up to that time Answer: It is often easier to specify a form for the hazard function than for the survival function (simpler math, more direct interpretation, flexibility in modeling)

If we wish to estimate the influence of factors of interest (e.g. body mass or gender) upon survival, how might we do so?

(1) Log-rank or binomial test (section 15.4, 3.1) - hypothesis testing non-predictive form (2) Model with linear-logistic regression - relies on specific parametric form (3) Proportion hazards or Cox model

Band recovery models presented today have a similar data structure and statistical methodology to the MRR capture models that will be presented on Friday. These MRR methods are often referred to as CJS models. What does CJS stand for?

Cormack-Jolly-Seber

Band recovery estimates of survivorship rely upon measurements of at least 4 different parameters to make their estimates. Name two of these parameters that are measured in these studies.

(1) k = number of release periods (2) I = number of recovery periods (3) Ri = number of animals banded and released during release period i (4) mij = number of animals banded during release period i and recovered during recovery period j

Band recovery data is often (see many tables in Chapter 16) expressed as a matrix of encounter frequencies. However, this matrix is typically not square (*hint: due to the sequential nature of data collection*). What shape do these matrices typically take?

Triangle

In these CSJ MRR studies, we are estimating "apparent" or "local" survival. What distinguishes this from Si or "true survival"?

Consideration of emigration *apparent/local: accounts for probability that an individual remains within a specific study area *true: accounts for both mortality and movements into or out of the study area

A number of different functions (identity, logit, logarithm, hazard) can be used to account for covariates when estimating survivorship. These are often (e.g. in section 17.1.4) referred to as ____ functions.

Link

Why are survival estimates from open population models (CSJ MRR models) less precise than those from closed population models?

They allow for migration, recruitment, etc. AND thus must estimate parameters for these attributes **MORE VARIABLES/PARAMETERS

How do the data used for Jolly-Seber models (Ch. 18) of abundance and recruitment differ form the data collected for CJS (Ch. 17) models of survivorship?

They do NOT expect you must quantify unmarked animals that are re-sighted

What does the Jolly-Seber model accomplish by adding explicit consideration of unmarked animals to the CJS approach?

It allows estimation of population size and recruitment

A critical distinction in the assumptions b/w CJS models and the Jolly-Seber approach is that in the Jolly-Seber approach you must assume that _____ probability parameters also apply to unmarked animals.

Capture

The robust design model discusses in today's lecture combines which two models from previous lectures into a single model for estimating survivorship and abundance?

Open and Closed

What limitations to open population models are addressed in the robust design through its combination of the two classes of models?

(1) Open model could not address heterogeneity among individuals (2) Open model could not address permanent trap response (3) Robust estimates of abundance (4) Robust design allows estimate of emigration (5) Robust design allows distinction between birth and immigration

Ch. 20 repeatedly mentions two authors of principle work in the robust design. One is named Pollock and the other Kendal. Which author developed the "ad hoc" version of the model and which the "MLE" method?

Pollock- Ad Hoc Kendel- MLE

In the description of the scientific method from the text, which form of causation (sufficient or necessary) specifies that the cause "C" under consideration alone insures occurrence of the effect "E"?

sufficient Sufficient Causation: The presence of the cause is enough to ensure the occurrence of the effect. Necessary Causation: The cause is required for the effect to occur, but its presence alone does not guarantee the effect.

According to the authors of the text, what are two keys to a successful model?

(1) Clearly define model objectives - predictive? mechanistic? compare scenarios? - who will be using it? (2) Include in the model only those features necessary to model objectives *Remember, models are only meant to be abstractions (simplifying reality to make it more manageable and understandable)

Why can't the method of moments technique for estimating probability questions about populations of interest be used to assess the reliability of estimates and perform tests of statistical significance?

Because this approach lacks information on the underlying distribution functions and THUS all information comes from the sample

Why would one want to modify the basic exponential model to allow for periodic collapse?

Real systems have limits, and thus populations crash when limits are exceeded

In Class Questions Flashcards

(80 cards)