Table 6.
Populations | Year | Estimated selfing rate(s) | Log likelihood | Delta deviance | d.f. | P-value |
---|---|---|---|---|---|---|
All pop | 2012 | sMET-2012, sNONMET-2012 | −3366·1 | 6·5 | 1 | 0·01 |
2012 | s2012 | −3369·3 | ||||
All pop | 2013 | sMET-2013, sNONMET-2013 | −3739·8 | 14·3 | 1 | 0·0002 |
2013 | s2013 | −3746·9 | ||||
MET | 2012–2013 | sMET-2012, sMET-2013 | −3912·8 | 0·1 | 1 | 0·75 |
2012–2013 | sMET | −3912·9 | ||||
NONMET | 2012–2013 | sNONMET-2012, sNONMET-2013 | −3193·0 | 0·2 | 1 | 0·65 |
2012–2013 | sNONMET | −3193·2 | ||||
All pop | 2012–2013 | sMET, sNONMET | −7106·1 | 21·7 | 1 | <0·0001 |
2012–2013 | s | −7116·9 | ||||
SB | 2010–2011–2012–2013 | sSB-2010, sSB-2011, sSB-2012, sSB-2013 | −2014·8 | 2·4 | 3 | 0·5 |
2010–2011–2012–2013 | sSB | −2016·0 |
The models were fitted using maximum log likelihood computed in RMES, and compared using log likelihood ratio tests.
MET, metallicolous; NONMET, non-metallicolous; pop, population; s represents the constrained or unconstrained selfing rate estimated for one population or a group of populations.
The best models are in bold, and the value of underlined estimated selfing rates are provided in Table 7.