Express the opportunity costs of you doing this kind of work.
It's hard to accomplish, and given your role as the least senior person in your role in a small company I think it might not be realistic for you right now.
I work in a similar role, and it's unfortunate but something like 60%-70% of applied statistical and data science work is data preparation and cleaning. And while it isn't difficult to perform, it is excruciatingly boring yet requires a fair amount of attention to do correctly. It's a perfect recipe for a numb mind.
The problem lies in the fact that there likely is not enough of this work to be done to justify hiring someone specifically to do it (like an assistant for you, personally). The task can be farmed out to someone else, but then it's hard to prioritize against other work required of whatever person gets stuck with the task.
Your current position seems to be:
I'm too good for this work, therefore I shouldn't have to do it.
That may or may not be true, but unless you are senior enough to command the resources to satisfy that second clause you'll have trouble getting what you want. Jobs are not about satisfying your personal preferences and desires, but about producing what you're told to produce. If you truly can't handle this, then many jobs in this field simply aren't for you and will not become so just because you dislike it.
Getting an assistant:
You can make a pitch to your boss about the relative value of your doing the mundane data tasks versus something else, but you'll need to have strong examples of what those other tasks are. If it's a small company there may not be a continuous stream of research projects for you to carry out (especially if you're not going to be doing the requisite data work) to justify having you on full-time. Your time on another task isn't more valuable because you won't be as bored, it's more valuable because the return on your time is greater than that of the dull task. It's not about this particular task being low-skill or low-value.
So you have to explain how much more the employer will be getting out of you working on the more interesting tasks. This can be hard with research, as outcomes are uncertain and it can be difficult to get a manager to appreciate the value of something unexciting (like a null result, or concluding that their pet idea is impossible).
I consider (and my boss agrees with me on this) that the dull, non-specialized tasks I do (like report preparation) pay for my research activities. My coworkers and superiors request specific reports, and so obviously at least perceive value in them, which gives me space to work on things where the value is less obvious to them. This is especially true of professional development, like reading up on and practicing new statistical techniques or investigating quirks and limitations of different approaches, which are crucial to working in this field.
The key to this argument is going to be a precise explanation of what you will do instead of report generation, and why that will be more valuable to the company than the report generation.
It's not the most pleasant prospect, but you can pay your dues in a less-senior position dealing with work your superiors can foist off on you until you achieve a more senior position of your own, and can then do the same yourself. Doing this isn't a total waste-- you may find you learn a lot about how to build data sets for analysis, how to work with arbitrary limitations in your company's database architecture, or other things. But, as above, the major objective of this kind of work is not personal satisfaction or stimulation, and your employer will not view it that way either.
You could also look for another job elsewhere, in a position that won't require this kind of thing or in a department that has the resources to delegate these tasks immediately. At the extreme end of this you could start your own business, such as a consultancy, and staff it as you like while directing work as you feel is most appropriate.
The best option for you:
Figure out how to write code that automates generating this report. It costs the company nothing, you'll learn a lot from doing it, and not only will this report no longer be a bother for you but you'll be better equipped to handle future boring, repetitive tasks. You can do a lot to manipulate data in a variety of programs. R probably isn't ideal for this, but can do it. Excel may also work and is likely to be available to you.
Even if you can't automate the entire thing, you may be able to make portions of the task easier on yourself.