Steve Tippins explains when you can say "I don't know" at a dissertation defense: only for questions outside your research scope. For anything touching your methodology, statistics, or data, you need to know it cold.
If I asked a question in a dissertation defense and somebody said, I don't know, that's a legitimate answer. If my question is outside of the boundary of what they've done, hey, what would happen if we did this on a different population? It's easy to say, I don't know because it's a different population. If I ask, what's the underlying assumption to the distribution of the variables that you've collected? And they say, I don't know, that's something they need to know. If it's within the bounds of their research, within their methodology, things like that, you know, you're showing your committee, warranting to your committee that you did this work and you understand it, you need to be able to say, you know, answer questions about it.
