This sounds like a HW question, so you need to edit your question to contain your argument for finding answers 3 and 4 correct, so that instead of just giving you answers, you are forced to defend your reasoning and learn from the question. This also seems to be a duplicate post here.
It would help to keep in mind that deductive inference is that which has a certain or deterministic conclusion, and inductive inference is one that is uncertain or probabilistic. Remember that an inference is a process by which one moves from premises to conclusion, and that premises themselves often have meaning grounded in unstated assumptions. Whether one's inference is deductive, inductive, or abductive, the question of not whether new information is created is largely a function of how one defines information. For instance, difference forms of statements might be considered novel forms of information to one who parses syntax as opposed to one who parses semantics.
EDITS: I see by your comments below that this is a hypothetical quiz question, and it's relation to the other post.
Inductive inferences rearrange current knowledge in such a way that they merely explicate what we already know.
Inductive inferences are such that if they are valid, the conclusion must be true if the premises are true.
Inductive inferences go beyond information we already have, thus they amplify our knowledge when they are used.
Inductive inferences are typically based on unfounded assumptions.
I'm not going to answer your questions directly, because in a way, since you wrote the questions, only you can really say what the answers are with certainty; however, some food for thought.
Take two examples, induction first. If a locust doesn't emerge for 6 seasons in a row, but then comes forth on the 7th cyclically, here's a perfect example of how after 6 years, one could claim based on the prior 6 years, they won't come around on the 7th, only to be surprised. This is in stark contrast to a man in the kitchen who is in the house who has to be in the house if he is in the kitchen. Knowing he is in the kitchen, we have the ability to state after inference he is in the house with certainty.
Before we begin, if we use justified, true belief as a basis of knowledge, we can add another dimension of analysis. So, let's look at an example of the inference.
- Periodic cicadas come out sometime to reproduce and carry on the genetic line.
- They don't come out every year.
- It has been 12 years since they have been out last.
Conclusion: They will come out the 13th year.
Here certainly we can make the case that our three premises are empirically and rationally valid, but how did we get to the conclusion? It's an inference, but it doesn't seem to be strong. How do you tell in an inference if your assumptions are unfounded? Does it rely on your definition of unfounded which seems tied up in your theory of truth?
- Periodic cicadas come out sometime to reproduce and carry on the genetic line.
- They don't come out every year.
- It has been 12 years since they have been out last.
- The last time they came out, it was on the 13th year, but sometimes they come out after 17 years.
Conclusion: They will come out the 13th year.
Note, that premise 4 seems to add strength to our induction, but it's results are still not certain. Now, remember that if knowledge is justified, true belief, in a way, we have added no knowledge at all with our prediction, because even though we have justified our belief, we don't know yet if it's true. So based on these grounds, it's hard to the conclusion of an induction ever really adds knowledge, but rather functions as a prognosticator. Remember that knowledge is meant to express the certainty of truth of the current or past state of affairs, where a prediction is meant to state certainty about the future. Let's say the cicadas emerge on the 13th year as predicted. Do we know (that is do we have true, justified belief) because of our inference, or is our empirical observation the ultimate source of truth and justification for the prediction? The epistemologist Robert Audi believes that there are five sources of knowledge, perception, memory, consciousness, reason, and testimony. If that is so, how do we disentangle reason from memory or perception generally? Does the act of the inference presuppose memory or perception?
As for new knowledge, let's say our induction introduces the conclusion as a claim in a chain of inference. Let's say that we presuming there is a modality of likelihood to the truth of our claim. Is the presupposition of the claim's truth in the next inference really the grounding for it becoming knowledge in the argument, and isn't that itself an assumption or inference warranted by some other part of the argument?
Ultimately, I would say that the stronger your arguments for each of the elements, the more likely they are to be considered correct by your peers. It is this process of justification that makes 'correct answer' mean 'true' among philosophers.
PS As a practical note, sometimes in science and statistics, the null hypothesis is used (https://en.wikipedia.org/wiki/Null_hypothesis) wherein a falsity is presumed true and then rejected statistically. Think about this in relation to your #4.
