QUIZ 2 Flashcards
1
Q
discrete variables
(3)
A
dichotomous = binary (female/male)
categorical = nominal ( bus, bike, walk)
ordinal ( very satisfied –> neutral –> dissatisfied)
2
Q
interval = numerical data
(i.e. differences are meaningful)
3
A
count
ratio ( height)
continuous (time)
3
Q
measurement
A
precision vs accuracy
significant figures
4
Q
describing distributions
A
univariate or bivariate
5
Q
univariate
5 characteristics
A
normal
uniform
bimodal
U-shaped
skewed
6
Q
descriptive stats
3
A
measures of central tendency
measures of variation
shape ( higher moments)
7
Q
measures of central tendency
3
A
mean, median, mode
8
Q
measures of variation
5
A
range, variance and standard deviation, quantiles, percentiles, inter-quartile
9
Q
shape (higher moments)
3
A
variance, skewness, kurtosis
10
Q
confidence
shown by 3
A
standard error
confidence interval ( credible interval)
statistical significance
scientific/ economic/ clinical significance
11
Q
likert scale example
A
strongly disagree–> disagree–> neutral –> agree –> strongly agree
( this data is categorical, ordinal, dichotomous, continuous)
12
Q
accuracy
A
true. consistent with the truth or objective (unbiased)
13
Q
precise
A
detailed, specific, having low uncertainty, highly resolved
14
Q
measurement e.g. the flood water is 35 cm deep, plus or minus half a cm
A
when we measure something in science, we always provide an estimate of our confidence along with the measurement
15
Q
estimation e.g. ospreys live on average 38.4 +/- 5 years
A
when we statistically estimate something that was not measured directly, or infer something about the world through our research methods, we must also calculate a measure of confidence for what we report
16
Q
communicating precision = probabilistic statements
A
between 31% and 37% with 95 % confidence ( i.e., e [31%,37%] 19 times out of 20)
17
Q
displaying univariate qualitative (categorical) data
3
A
tables, bar charts, pie charts
18
Q
displaying univariate continuous data
3
A
histogram, box plot, kernel density estimation (carpet plot)
19
Q
bivariate plots: two categorical variable
1
A
paired bar plots
20
Q
how to visualize –> bivariate plots : one categorical, one continuous variable
2
A
multiple histograms, box plots
21
Q
visualize –> bivariate plots: two continuous variables
1
A
scatter plots
22
Q
univariate measures of dispersion
5
A
range
standard deviation: average distance from the mean, coefficient of variation, index of dispersion, interquartile range
23
Q
measures of shape
4
A
variance, skewness, kurtossis, L-moments
24
Q
Skewness Left (negative), Right (positive)
A
(a) negative direction ( mean - median - mode)
(b) positive direction ( mode -median-mean)
25
confidence interval
the range within which we would expect the value of the statistics to fall, if we were to repeat the study with a very large sample
26
what does confidence depend on?
sample spread, sample size, ( the nature of the statistic you're estimating )
27
credible intervals are a simpler concept from bayesian stats
the interval within which an unobserved parameter value falls with a particular probability, given available data, model, and preexisting knowledge
- best estimate of the true value of parameter
28
properties of a univariate variable
ex: describe (aspects of) the distribution of an unordered list of sample measurements
29
multivariate properties
-two things measured for each individual sampled
- describes (aspects of) the relationship between theme
30
univariate measures of central tendency
mean, median, mode
31
statistical significance
"within 3%, 19 times out of 20"
- closely related to confidence intervals = error bars
| observed relationship is unlikely to be due to chance
32
scientific ( or economic) significance
an estimate of effect size put into context, for instance by competing effects, natural variation, or comparison to other costs or impacts
33
discrete variables
dichotomous = binary (female/male)
categorical = nominal ( bus, bike, walk)
ordinal ( very satisfied --> neutral --> dissatisfied)
34
# Determine H & H0.
Are these outcomes equally likely?
H: are the counts /frq. of each category as expected?
H0: the probability of each group is exactly equal (or equal)
35
# Determine H & H0.
are groups different from each other?
H: are the means of X different in the two groups
H0: the mean of Xa equals the mean of Xb
36
# Determine H & H0.
did the outcome change?
H:did the mean of outcome X change across the two measurements?
H0: the mean of X1 equals the mean of X2
37
# Determine H & H0.
are these two outcomes correlated?
H: are they the same?
H0: X and Y are uncorrelated (orthogonal)
38
# Determine H & H0.
are these variables related
H: is there a linear relationship between X and Y
H0: the coefficient of alpha in an OLS regression of Y on X (and Z...) is zero
39
Procedure for a statistical test
1-based on theory + H0 there is a stat of interest that you can calculate with sample
2-use theory to derive distribution of expected values (under H0). some assumption must be made
3- calculate the statistic actual value given sample
4-state likelihood of answer from sampling method, given that the H0 is true, if unlikely = reject H0
40
What is the null hypothesis (H0)?
-what you are trying to disprove
- strong results, eliminate the possibility of the H0
- no difference, no change, small difference, no effect
41
reject the null if ( statistical significance)
p-value is small or near zero
42
type I error for binary decisions
rejecting the null hypothesis when it is true ( false positive)
43
type ii error
failing to reject the null hypothesis when it is false (false negative)
44
the probability of a type I error is determined by ____
significance level
- with a 99% significance threshold, type I error is less likely than with a 95% confidence
45
the probability of a type ii can be computed for a particular test statistic (i.e. H0), if given______
population distribution (parameterization, mean and SD)
sample size (N) and alpha ( chance of type I error)
46
hypothesis test power
1 - ( the probability of a type ii error)
AKA - the probability that we correctly reject the null, when the null is false
47
frequentist statistics
developed before computers & calculators ( t-test, chi-square, F test)
48
bayesian statistics
rapidly developing framework for using prior expectation + evidence to make bets
49
measurement validity
how well your metric captures the underlying concept you are trying to measure
50
internal validity
the degree to which the design of an experiment controls extraneous variables
51
external validity
the degree to which effects found in an experiment generalize to other individuals, contexts, and outcomes
(ex: in sampling studies, are the times and places outside of the sampling frame represented accurately)
52
# Threats to external validity
interaction of selection and treatment
(unrepresentative responsiveness of treated pop.)
53
# Threats to external validity
interaction of setting and treatment
effect of the treatment may differ across geographic or institutional settings
54
# Threats to external validity
interaction of history and treatment
effect of the treatment may differ across time periods
55
# Threats to external validity
effect may not persist
-as individuals and institutions adapt over time to the treatment
56
# Threats to external validity
Partial- Equilibrium Effect
other components of the system also undergo related changes, reducing or eliminating the effect
57
# measuring biodiversity
total species list
most useful for well-studied groups such as birds and large mammals
58
2 communities in which total counts are not feasibile
soil microbes + coral reefs
| too diverse to count
59
# measuring biodiversity
subsample using quadrats or transects
count all species on several quadrats or transects of known size and extrapolate the average to the whole area
60
reliability of quadrats depends on (3)
1- # of species in each quadrat must be determines exactly
2- area of quadrat known
3- quadrat counted must be reprsentative of the the whole area
| some form of random sampling
61
Quadrat sampling can be ___
| type of os sampling (4)
random, systematic, stratified random or snowball sampling
62
# measuring biodiversity
time-restricted search // rapid biodiversity assessment
habitat is searched for a set period of time & species recorded
63
# measuring biodiversity
species discovery curves
construct a species accumulation curve
| highly depedent on sampling effort ( time invested + ground covered)
64
Quantifying density across transects
-relative density
-camera traps
-tracking stations
65
capture-recapture methods
capture,mark, release
- in subsequent samples, the portion marked should be representative of proportion marked in whole population
66
biodiversity indicators
an indicator species'status reflects the status of other species in a community
67
ex: taxon based indicator species
woodpeckers are indicators of forest bird diversity
68
rapid inventories for conservation
conservation tool designed to assess the biological importance of priority sites for conservation
69
conservation priority sites
plants, birds, mammals, fishes, frogs, geology, soil diversity, water characteristics
HUMAN DIMENSIONS
70
observational methods for behaviour sampling
ad lib
focal animal
scan sample
one-zero sampling
71
# method descriptions
focal animal (pair or group)
restricts data recording during sample period to one animal, pair, or group
72
# method descriptions
ad-lib
opportunistic sampling with no constraints
73
# method descriptions
one-zero
record the occurence (1) or nonoccurence (0) of selected behaviours during sequential sample intervals
74
# method descriptions
instantaneous or scan
record behaviour of an individual (instantaneous) or group (scan) at sequential, predetermined points in time
75
selection of animal
- random
- systematic
- stratified
76
indirect measures
- stress hormones
- ketones
- c-peptide
77
# habitat selection
how organisms evaluate + select an environment from the different alternatives available
non-random distribution
| *behavioural component (choice)*
78
how to determine habitats
correlative studies
- assume species density is proportional to the quality of habitat
| Problem!!! Source - Sinks
79
ideal free distribution theory
competing individuals distribute themselves between reosurce patches in such a way that each individual receives the same payoffs
80
ideal free distribution theory assumptions
| equality
- first animals use the best patch
- no difference in competitive ability
- no territoriality
81
ideal despotic distribution theory
| reality?
- competitors exclude others w/ aggression
- 1st arrivals establish in high quality habitats + defend resources
- lower quality habitats fill up with later arrivals
- if full, future arrivals must take over or be excluded
- fewer competitors and higher intake rates in richer habitats
82
What is ethics?
study of general nature of morals and moral choices
| morals: conforming to established ideas of right and wrong
## Footnote
*unwritten guidelines*
83
ethics vs laws
| laws are....
formal, written standards
- designed to apply to everyone
- enforced by gov agencies
- interpreted by courts
| impossible to pass laws to cover every possibility
84
unethical behaviour
- isn't necessarily illegal
- not all illegal behaviour is unethical
- not conforming to approved standards
85
amoral behaviour
- no sense of right or wrong
- no interest in moral consequences
86
which system of ethics works best?
- no universal agreement
- ethical decisions are greatly influenced by personal ethics
87
# McGill & Unethical research
Project MK Ultra
- CIA mind control experiment
- unconsenting patients
- sensory deprivation, LSD, electroshock therapy, and other mind control methods
| devastating effects on patients invoved
88
Research ethics board at McGill
proper certification required before research if working with
- human participants, animals, radioactive or biohazardous materials
89
Enforcing ethics
- no data can be collected without or prior to an approval
- funding withheld until approval
90
General rule of ethics requirements
**always** required with children
assume any human subjects
| EXCEPTION--> designated representatives of organizations
91
Research ethics board-1 (REB-1)
law, arts (except: social work, information studies), engineering, management, continuing studies, religious studies, science (except : psych)
92
REB-2
linguistics, psych, music, social work, information studies, education + research involving competent adults
93
REB -3
all faculties (except med and dentistry) involving minor or uncompetent adults (unable to consent)
94
REB-4
agricultural and environmental sciences involving competent adults
95
# application process
Required human research ethics training
tri-council policy statement 2
| all students, faculty, staff, investigators
96
# application process ii
who can apply as a prinicpal investigator?
NO undergrad, profs can apply for students (if human participants is needed)
97
# submission deadlines
minimal risk
- submitted at any time
- review ~4-6 weeks
98
# submission deadlines
projects with greater risk +involve minors
- 1st friday of every month
- REB meets 4 weeks after that
- no board meetings in july /aug
99
informed consent
- explicitly obtained (written or verbal)
- voluntarily given
- can off compensation BUT must be mentioned +approved by REB
100
confidentiality
- identities or sensitive info can not be revealed in publication
- all data must be securely stored (paper+ ecopy)
- kept for 7 years
- specific ways of getting rid of data
101
# confidentiality example
Poaching traps in Kibale National Park
- serious repercussions for locals who hunt for subsisdy or artisanal trade
| minimize legal repercussions
102
Why overfishing an interdisciplinary problem?
- fish (species & abundance)/ ecology
- managerial practices (fishing/econ)
- regulations (institutionalization of a field)/ politics
103
lake victoria's fisheries
- dynamic eco-history of the lake victoria basin
- high population density
- supports Africa's largest inland fishery
104
history of fishing in the area
- intensified with new tech
- intro of new species influenced faunal compositon & balance
105
fish that flourished
| even with gigantic nile perch "mputa"
- nile tilapia "ngege" ( has become abundant, commercially important )
- "mukene", biomass 6x
106
# *changes*
lake victoria system composition
multi-species system exploting native fish --> 3 species almost all catch ( 2 are non-native)
107
Similar changes have occured in which other lakes in the victoria basin?
Lake Nabugabo
Kyoga
108
Changes in perch population
- increased fishing
- decrease in stock
- decrease in size
109
lake ecology
| not homogeneous
understanding habitat diversity is key to understand relative fish abundance and reproduction
110
ecological morphology
refers to resource availability (oxygen, nutrients, protection etc..)
111
Biodiversity Banks (3)
satellite lakes, rcky refugia, wetland refugia
112
Beach management Units
promising solution/control of socioecological system
113
BMUs structure
new institutions that regulate fishing in the lakes
- to fish, fisherman must register
- control gear used
- no territorial jurisdiction
- no jurisdiction over capture (type or quantity)
114
Overfishing is class case of ...?
tragedy of the commons or open access resources
115
Collective Action Problem (CAT)
predicts a resource managed in common has more probability to be sustainably managed with certain characteristics
116
increasing probability of sustainable management
- small size (group +resources)
- well-defined boundaries
- high level of dependence on resource
- fair allocation
- locally devised access and rules
- clear territoriality
- resource harvest restrictions match recovery
117
flaws of BMU design
| ID by CAT predictors
- lack of territorial jurisdiction
- no regulations on harvest
- fisheries highly unpredictables (spatial distribution, compositition..)
| BMUs = insufficient institutional step
118
Why are fisheries so hard to manage?
- difficult to institutionalize, and generate territorial jurisdictions?
119
Data required to manage fisheries?
- ecological distributions of resources (ecological map, fish presence+mobility, ethnoecological mapping)
- spatial distributions of the fishing effort
- trade networks
120
How to create a map of ecological structure?
| in lake nabugabo
- ecological structure (depth+ shoreline ecotones: lilies, grass, forest edge)
- quantify fish movement (tag)
- etnoecological mapping (cognitive map +interviews)
121
Collecting data on fishing effort
| in lake Nabugabo
- use ecological map as base, recall interview designs to determine where BMU members fished during 5 days of each month
122
example of fishing recall qs
- where did you fish today? how long? why there?
123
catch landing surveys
both BMUs
boats, type of fish & gear
124
expected results
| Lake Nabugabo
define specific ecological patches with significant fish species distribution ==> define exclusive territorial jurisdictions for the BMUs*
| *worth defending*
125
consequences of no predictable pattern
| in fishing efforts
- cannot design functional institutional structure
- no incentive for collective action
