basic stuff

Question 1

Direct, long gradient

Answer 1

CCA

Question 2

Direct, short gradient

Answer 2

RDA

Question 3

Constrained

Answer 3

CCA and RDA

Question 4

unconstrained

Answer 4

CA and PCA, distance based approaches (mds and polar)

Question 5

Indirect, long gradient

Answer 5

CA

Question 6

Indirect, short gradient

Answer 6

PCA

Question 7

Parametric (linear) distance based approach

Answer 7

Classical MDS

Question 8

Non-parametric (ranks) distance based approach

Answer 8

nonmetric MDS

Question 9

when do you add environmental data afterwards?

Answer 9

indirect ordination

Question 10

long gradient response?

Answer 10

unimodal

Question 11

short gradient response?

Answer 11

linear

Question 12

What do Log (1+x) and sqrt(x) transformations do?

Answer 12

Reduce the influence of large values

Question 13

What does the Hellinger transformation do?

Answer 13

computing sqrt of relative abundances per location

Question 14

What does the Wisconsin transformation do?

Answer 14

‘double standardization’: first computing relative
abundances per location, then normalizing values to max
value=1 per species

Question 15

which distances are symmetric?

Answer 15

Manhattan and Euclidian

Question 16

which distances are asymmetric?

Answer 16

chi-squared and Bray-Curtis

Question 17

what is inertia for CCA/CA

Answer 17

Chi-square

Question 18

what is inertia for RDA/PCA

Answer 18

Variance

Question 19

what distance is CA based on?

Answer 19

Chi-squared distance

Question 20

what is the average MDS by design?

Answer 20

0

Question 21

when is stress level acceptable?

Answer 21

<0.05 is excellent, <0.1 is great, <0.2 is good/ok, <0.3 is poor

Question 22

how are distances mapped in Non parametric MDS?

Answer 22

The ranks are mapped

Question 23

is non paramteric MDS the same result every time you try?

Answer 23

No, it’s iterative (each run may have different results)

Question 24

how to check the quality of NMDS?

Answer 24

Inspecting stress value, Shepard diagram

Question 25

how to choose how many dimensions for NDMS?

Answer 25

The least amount of dimensions where the stress is still below 0.1 (more dimensions reduce stress but you want the least amount of dimensions)

Question 26

how to choose how many axis for PCA?

Answer 26

scree plot with broken stick method

Question 27

what is the total inertia for CA?

Answer 27

Average chi-squared distance among all sites, or sum of all eigenvalues

Question 28

what do eigenvalues represent for CA?

Answer 28

the chi-square value per axis

Question 29

what distance is PCA based on?

Answer 29

Linear species response based on Euclidian distance

Question 30

when is a gradient long or short?

Answer 30

first axis of a DCA (detrended) is 3 or <3 it is short. Also if there are many zeroes it is most likely long, and most species in most sites means short.

Question 31

how to know in output if a test was constrained (CCA/RDA) or not (CA/PCA)?

Answer 31

if output does not say 'constrained' it is unconstrained

Question 32

how to calculate proportion explained by an axis?

Answer 32

eigenvalue of axis / total inertia

Question 33

in a significance test for constrained ordination, what refers to restrained and unrestrained part?

Answer 33

residual= unrestrained. Model = restrained

Question 34

when is inertia chi-square?

Answer 34

for CA/CCA

Question 35

when is inertia variance?

Answer 35

PCA/RDA

Question 36

what is total inertia for PCA?

Answer 36

sum of the eigenvalues of all axes, or sum of all variance (since inertia = variance)

Question 37

what does it mean when the outcome of an unconstrained and constrained analysis are very different?

Answer 37

the enviornmental variables do not explain the variance well

Question 38

when can you use multivariate analysis

Answer 38

for anything that has multi-response data that allowes calculating dissimilarities

Question 39

in an constrained output how many axes are used?

Answer 39

the constrained + unconstrained ranks. constrained = amount of environmental variables

Question 40

when do you have to check for colinearity of explanatory variables?

Answer 40

with CCA and RDA because they are regression methods

Question 41

when is collinearity a problem?

Answer 41

vif >5 potentially, >10 definitetly

Question 42

what is conditioning?

Answer 42

removing the effect of gradients in a landscape for further testing

Question 43

what are canonical axes?

Answer 43

constrained axes

Question 44

what is a marginal effect?

Answer 44

the effect of a particular term when all other model terms are included in the model

Question 45

how are marginal terms tested?

Answer 45

by eliminating each term from the model containing all other terms

Question 46

are marginated effects dependent on order of the terms?

Answer 46

no, but correlated terms will get high P values

Question 47

does the order matter when testing terms?

Answer 47

yes, if there is correlation

Question 48

what is the P value in a permutation test?

Answer 48

the fraction of permutation values that are larger than the actual value

Question 49

how is p value calculated in permutation test?

Answer 49

the number of values that are higher than the actual value, divided by (the amount of permutations + 1)

Question 50

in an unconstrained ordination do the eigenvalues increase or decrease over the axes?

Answer 50

decrease

Question 51

when do you use chi-square distance

Answer 51

when long gradient and rare species are well sampled

Question 52

when do you use Euclidian or mannhattan distance

Answer 52

when there's a short gradient

Question 53

when do you use bray-curtis disctance

Answer 53

long gradient and rare species not well sampled

Question 54

what to be careful of with Manhattan/euclidian distance?

Answer 54

Scaling, double zeroes, abundance paradox

Question 55

what to be careful of with Bray curtis distance?

Answer 55

Large values

Question 56

what to be careful of with Chi square distance?

Answer 56

rare species (that aren't sampled well), distance between two sites is influenced by other sites