Advice from all who have completed an MS Thesis

[random_soldier]

Do you think you can sum up how it is supposed to go? The entire process?

 

[Aufbauwerk 2045]

Without going into details, in my case it was very simple. I got hired as a research assistant by a professor who needed help with a project he was working on. I spoke to him before I even applied to that school. He liked me and recommended me for admission.

He did most of the research work. My job was to check his work and catch any errors if I could. Also I implemented his ideas in a computer program. Based on this program, he was able to publish a paper in some conference proceedings. He added my name to the paper. I always thought my contribution was not deserving of having my name on the paper as a coauthor, but he disagreed.

I began work on my thesis during my second year. I explained the theory and also how I implemented it in a program. The thesis was accepted. The whole thing was quite easy. The only downside for me was that I needed to write the thesis on his topic, instead of something I selected on my own. But I did not argue, since I wanted to finish the M.S. as soon as possible. Also I figured I owed him, since he was kind to me and I think his recommendation helped me get into the graduate program. So I did the best I could on his project.

This was my rather easy road to a M.S. I look forward to hearing if other people had a different sort of experience.

P.S. Of course when I say "easy" I should mention I worked 20 hours per week as a research assistant or a teaching assistant, while taking a very heavy course load of STEM subjects, and later having to dedicate many hours to writing my thesis. What was easy about it was not my work load, which was intense, and left almost no free time for two years. It was easy in the sense that I was following a definite path towards getting my M.S. It was made easier by the fact that the professor did not try to exploit me in any way by dragging things out. I was treated quite fairly by him and by the others on the thesis review committee.

 

[PhanthomJay]

I did my thesis on a pet project of my professor , at his request, which expanded a prior thesis on one story building frames to my thesis on multi-storied frames, using finite difference hand analysis. I too wasn't much interested in this project, but its what the prof suggested so I went along with it and although it took a lot of time, it wasn't overly difficult because I used the prior thesis as a guide, not much interaction was involved, and I took the structural analysis course first, given by that professor on the same topic, in the prior quarter. I then made an oral presentation to the thesis committee, answered some questions , some right, some wrong, and the thesis was approved as well as the degree. I don't think all schools require a thesis for the MS...I got mine 50 years ago (my prof just passed away last month). I completed my MS in one year, the thesis was during the final quarter, and the first three quarters were the hardest because of the intensity of the advanced courses.

 

[random_soldier]

Without going into details, in my case it was very simple. I got hired as a research assistant by a professor who needed help with a project he was working on. I spoke to him before I even applied to that school. He liked me and recommended me for admission.

He did most of the research work. My job was to check his work and catch any errors if I could. Also I implemented his ideas in a computer program. Based on this program, he was able to publish a paper in some conference proceedings. He added my name to the paper. I always thought my contribution was not deserving of having my name on the paper as a coauthor, but he disagreed.

I began work on my thesis during my second year. I explained the theory and also how I implemented it in a program. The thesis was accepted. The whole thing was quite easy. The only downside for me was that I needed to write the thesis on his topic, instead of something I selected on my own. But I did not argue, since I wanted to finish the M.S. as soon as possible. Also I figured I owed him, since he was kind to me and I think his recommendation helped me get into the graduate program. So I did the best I could on his project.

This was my rather easy road to a M.S. I look forward to hearing if other people had a different sort of experience.

P.S. Of course when I say "easy" I should mention I worked 20 hours per week as a research assistant or a teaching assistant, while taking a very heavy course load of STEM subjects, and later having to dedicate many hours to writing my thesis. What was easy about it was not my work load, which was intense, and left almost no free time for two years. It was easy in the sense that I was following a definite path towards getting my M.S. It was made easier by the fact that the professor did not try to exploit me in any way by dragging things out. I was treated quite fairly by him and by the others on the thesis review committee.

When you say your professor did most of the research work, do you mean to say that he designed the work to be carried out or that he personally collected the data?

Also, is that allowed? You basically explaining the theory of what you worked on with your professor and how you carried out the experiment? I thought original research was a requirement.

Finally, you say you did your best on his project. I'm confused as to the meaning since you already worked with him on something that you used for your thesis. Did you do another project with him? Was it compulsory?

I did my thesis on a pet project of my professor , at his request, which expanded a prior thesis on one story building frames to my thesis on multi-storied frames, using finite difference hand analysis. I too wasn't much interested in this project, but its what the prof suggested so I went along with it and although it took a lot of time, it wasn't overly difficult because I used the prior thesis as a guide, not much interaction was involved, and I took the structural analysis course first, given by that professor on the same topic, in the prior quarter. I then made an oral presentation to the thesis committee, answered some questions , some right, some wrong, and the thesis was approved as well as the degree. I don't think all schools require a thesis for the MS...I got mine 50 years ago (my prof just passed away last month). I completed my MS in one year, the thesis was during the final quarter, and the first three quarters were the hardest because of the intensity of the advanced courses.

How did you manage your MS in a year? Was it required to be done so? Sounds hectic.

 

[PhanthomJay]

How did you manage your MS in a year? Was it required to be done so? Sounds hectic.

It wasn't required, but I was anxious to get it done. It was 12 months (4 quarters) of study in advanced structural topics like Stability, Elasticity, Plates and Shells, and such. It was hard work, but I managed a B average. The actual degree was given a year later with the 2 yr program grads, but I had a letter stating that my degree was approved, so I got my first job prior to my actual MS degree, but my MS degree was accepted by the employer and it rewarded me a 10% higher starting salary than a BS degree , earning me a whopping $10 K/yr starting salary in 1968 dollars. Many schools in US have 1 year masters programs.

 

[random_soldier]

It wasn't required, but I was anxious to get it done. It was 12 months (4 quarters) of study in advanced structural topics like Stability, Elasticity, Plates and Shells, and such. It was hard work, but I managed a B average. The actual degree was given a year later with the 2 yr program grads, but I had a letter stating that my degree was approved, so I got my first job prior to my actual MS degree, but my MS degree was accepted by the employer and it rewarded me a 10% higher starting salary than a BS degree , earning me a whopping $10 K/yr starting salary in 1968 dollars. Many schools in US have 1 year masters programs.

I really respect that and I wish it were possible for more people to do that more often. Just get it over and done with.

Does the fee get reduced, though or is it still per credit-hour so you still have to pay the same regardless of the time spent?

 

[analogdesign]

In the US, graduate programs are charged on a quarterly or semester basis. There is also a "filing fee" status where you don't use university resources (for example, like PhanthomJay you are working but don't have the paper degree yet) where you pay a nominal fee to remain associated with the program. I completed the work for my PhD program and my advisor allowed me to write my thesis while I worked (since I was supporting a family... I am forever grateful for that). So, I paid something on the order of $100 a quarter for the three quarters I worked and wrote my thesis during nights and weekends.

Regarding the MS thesis, it is extremely rare an MS thesis is groundbreaking or even very original (Shannon's boolean logic thesis comes to mind as a big exception). Usually they are a deep dive into a known topic and maybe if you're lucky you can push the state-of-the-art a bit, but it is categorically different from a PhD thesis.

 

[onoturtle]

The process for me went as follows, as a CS major,

- Find an advisor. I initially was in an area about theorems/proofs (information theory). I got out of that after my first MS year to do more empirical work (applied machine learning). I then chatted with 3 professors, got rejected by one (advising too many students already) and eventually picked one. This is probably the most critical part of the process. You need to pick professor and an area of study that's going to work for you. Chat with the existing students to see if that prof style matches your's. Do you need a micromanager? Or someone that gives you a lot of space and freedom? Etc.

- Pick topic. I had no idea, so my advisor proposed one and I ran with it. The advisor had the outline of an approach. I tried it, part of it failed, and I came up with a different approach that ended up as the final approach. Before starting research, I think the university required a thesis proposal document, but no defense was required so it was just up to the advisor to approve it or not. In my case, the topic and approach was their idea, so it was more of a check on my understanding of the problem, background literature, etc.

- Do research. How this plays out depends on your advisor. My setup was monthly meetings where I report my progress. Discussion of any issues that came up via email and additional meetings if needed.

- Write about research. Probably should be doing this while doing the research rather than saving it all for the end. You'll have to deal with all the formatting issues that your university likely requires.

- Thesis defense. You put together a committee that your advisor chairs. They all supposedly read the thesis. You present the thesis with powerpoint presentation. You do Q&A and discuss with committee. Committee decides if you pass or not.

- That's the end of the program for me, but often I imagine your advisor would like you to prepare publications from the thesis. In my case, it was just one pub. Unfortunately, IBM Watson Labs beat me to the punch in the middle of my research so the impact of my work was lessened.

For me my MS took 2.5 years. Maybe if I didn't wipe out my research from the first year and start fresh in my second it would've taken the typical 2 years I was aiming for.

 

[random_soldier]

In the US, graduate programs are charged on a quarterly or semester basis. There is also a "filing fee" status where you don't use university resources (for example, like PhanthomJay you are working but don't have the paper degree yet) where you pay a nominal fee to remain associated with the program. I completed the work for my PhD program and my advisor allowed me to write my thesis while I worked (since I was supporting a family... I am forever grateful for that). So, I paid something on the order of $100 a quarter for the three quarters I worked and wrote my thesis during nights and weekends.

Regarding the MS thesis, it is extremely rare an MS thesis is groundbreaking or even very original (Shannon's boolean logic thesis comes to mind as a big exception). Usually they are a deep dive into a known topic and maybe if you're lucky you can push the state-of-the-art a bit, but it is categorically different from a PhD thesis.

Does the way you mentioned charging to work go the same for internationals?

The process for me went as follows, as a CS major,

- Find an advisor. I initially was in an area about theorems/proofs (information theory). I got out of that after my first MS year to do more empirical work (applied machine learning). I then chatted with 3 professors, got rejected by one (advising too many students already) and eventually picked one. This is probably the most critical part of the process. You need to pick professor and an area of study that's going to work for you. Chat with the existing students to see if that prof style matches your's. Do you need a micromanager? Or someone that gives you a lot of space and freedom? Etc.

- Pick topic. I had no idea, so my advisor proposed one and I ran with it. The advisor had the outline of an approach. I tried it, part of it failed, and I came up with a different approach that ended up as the final approach. Before starting research, I think the university required a thesis proposal document, but no defense was required so it was just up to the advisor to approve it or not. In my case, the topic and approach was their idea, so it was more of a check on my understanding of the problem, background literature, etc.

- Do research. How this plays out depends on your advisor. My setup was monthly meetings where I report my progress. Discussion of any issues that came up via email and additional meetings if needed.

- Write about research. Probably should be doing this while doing the research rather than saving it all for the end. You'll have to deal with all the formatting issues that your university likely requires.

- Thesis defense. You put together a committee that your advisor chairs. They all supposedly read the thesis. You present the thesis with powerpoint presentation. You do Q&A and discuss with committee. Committee decides if you pass or not.

- That's the end of the program for me, but often I imagine your advisor would like you to prepare publications from the thesis. In my case, it was just one pub. Unfortunately, IBM Watson Labs beat me to the punch in the middle of my research so the impact of my work was lessened.

For me my MS took 2.5 years. Maybe if I didn't wipe out my research from the first year and start fresh in my second it would've taken the typical 2 years I was aiming for.

Why was the importance lessened? I think there would be a fair bit for somebody else confirming that the experiment and the results can be repeated such that there are no discrepancies.

 

[Aufbauwerk 2045]

When you say your professor did most of the research work, do you mean to say that he designed the work to be carried out or that he personally collected the data?

Also, is that allowed? You basically explaining the theory of what you worked on with your professor and how you carried out the experiment? I thought original research was a requirement.

Finally, you say you did your best on his project. I'm confused as to the meaning since you already worked with him on something that you used for your thesis. Did you do another project with him? Was it compulsory?

There was no data at that stage. He came up with the theory. My job was to figure out how to implement his theory in a computer program, to write that program, and use it to produce computational data and graphics. Basically my work was computational physics, which was used to study the validity of his theory prior to any real world experiments.

All the computational work I did was totally original. I worked alone. Of course I had to demonstrate to the professor and the occasional moneybags visitor from industry the results of my work. There was grant money involved in this whole process. If you can find a way to help a professor reach an important milestone in getting more funding, then Bob's your uncle. Lucky for me, I satisfied everyone's expectations.

I call it his project because he was the director. I was merely his research assistant. I think that's the normal arrangement for students working on a thesis.

If it was "my" project in the sense of being totally self-directed, with no concern about grant money, and so on, then I would have chosen to work on my own pet idea of the time. But the system is set up so you get approval from your thesis advisor for your plan, and your finished thesis needs approval from a committee. Need I say more about the hierarchy involved?

That's all I can explain about it.

:)

 

[Aufbauwerk 2045]

It wasn't required, but I was anxious to get it done. It was 12 months (4 quarters) of study in advanced structural topics like Stability, Elasticity, Plates and Shells, and such. It was hard work, but I managed a B average. The actual degree was given a year later with the 2 yr program grads, but I had a letter stating that my degree was approved, so I got my first job prior to my actual MS degree, but my MS degree was accepted by the employer and it rewarded me a 10% higher starting salary than a BS degree , earning me a whopping $10 K/yr starting salary in 1968 dollars. Many schools in US have 1 year masters programs.

$10K a year in 1968 was not so bad. Taking inflation into account, that would be about $70K in today's money.

In 1968, median income was about $7700. The average house cost about $27,000.

I think in terms of actual living standards, you would have done far better making $10K in 1968 than $70K today.

 

[onoturtle]

Why was the importance lessened? I think there would be a fair bit for somebody else confirming that the experiment and the results can be repeated such that there are no discrepancies.

Being first is much better than being second. My work isn't a replication of their work (how could it when we started before that was published) and was independent of it so it took a different approach. And since we probably started work later, we could publish results on analysis of a use case they couldn't with at the time newly emerging next-gen sequencing technology. So there's was enough novelty there to justify the MS and later publication. Nevertheless, the novelty was lessened due to the existence of the other paper.

I agree on the point of replicating experiments and reproducible science. I've only very recently returned to academia since getting my PhD, so I'm not really sure what the current climate is, but publication of replicated experiments, null, negative, or inconclusive results isn't incentivized from what I know. If you want a bigger impact (more citations or whatever metrics), want to get into big journals, etc., that's not the way to go. There are venues to publishing such work (e.g. links below) and so perhaps could work out in the MS context, but that's not going to advance an academic career. That's arguably not how science should be done, but that's the current reality as I know it.

https://f1000research.com/channels/PRR
http://blogs.plos.org/everyone/2015...-focusing-negative-null-inconclusive-results/

 

[PhanthomJay]

$10K a year in 1968 was not so bad. Taking inflation into account, that would be about $70K in today's money.

In 1968, median income was about $7700. The average house cost about $27,000.

I think in terms of actual living standards, you would have done far better making $10K in 1968 than $70K today.

So true. And today's starting salaries are generally less than $70K. Bring back the '60s!