Start Studying Astrophysics or Uptake a PhD Offer

[Harshna]

Hi all,

I have completed a Bachelor of Science (Biological Sciences) and Graduate Diploma of Science (Applied Data Science). I have always loved Maths and Physics, but never really pursued it as I thought there were not many jobs for them (but I have taken first year Maths courses as electives and postgraduate statistics courses which were required for my Graduate Diploma). As my next step of study would probably be my last - I need to really consider what I am doing. I have realized now that I REALLY want to work in Astrophysics/Space Exploration. I also am interested in Machine Learning, but coding (including in Matlab) is not my strength. I have not taken any Physics courses at Uni, but have done Physics up to year 12.

I have two options for my next study:

1. A summer research scholarship with guaranteed progression to PhD with topic in "Predictive Modelling and Applications in Bio-physical Area with Data-intelligent Algorithms" where I would be using Machine Learning (Neural Networks) to model the Photosynthetic Active Radiation (Radiation of Sunlight at which growth is optimum) of algae from internet data sources
2. Pursue a Masters in Applied Data Science in which I have the opportunity to do 4 pure Astronomy courses (as the Applied bit) which will take 1 year part time (as I have credits) and then Masters by Research (2 years part time) in Astrophysics with Machine Learning (to allow me to have done enough research to meet the prerequisites of a PhD), and then do a PhD (3-4 years) in Astrophysics with Machine Learning. This option is in total another 6-7 years, but it is my dream

I love science and the PhD topic in Option 1 also interests me, but I think I will always regret not taking the Astrophysics option as Astrophysics is my new dream.

I have a few issues:

1. Will I get a job in Physics in Australia if I pursue the second option?? I know the Physics market is not big - so if I spend about 7 years and then get nowhere instead of taking the first option (which is much more in demand as it is environmental science which there are plenty of jobs for), would be a complete waste of time when I had a much better option
2. If I do the PhD topic in Option 1 (which involves Machine Learning but in Biology) than will I still have the opportunity to get a job in Astrophysics (I could possibly still take some Astrophysics courses later). I think companies would be looking more for Machine Learning skills for a job in Machine Learning in Physics
3. I am thinking of working (until I do a PhD at which time I will not probably work) so any qualification will take a longer time doing it part time with work
4. I am 23 years old so the second option may mean I end up finishing studies at about 30 years old
5. Approving a PhD topic and supervisor is not usually the easiest thing (especially in Physics when I haven't done Physics courses), but a Supervisor has contacted me himself and offered me the research scholarship with progression to PhD - so this is a really great opportunity and it is a much FASTER option (half the time)
6. If I take Option 2 I could possibly end up lending a job at my Uni after taking the PhD as they may give me a job on the same topic as my research (which could very much also happen for Option 1). What this means is that Option 2 actually may have a job opportunity, if the Uni lends me one.

I am so lost which option to choose as I know:

• Option 1 will get me a job, but I will always regret not doing Astrophysics when I could have
• Option 2 is my dream but I am very unsure as to whether it will quickly end up getting me a job or will be a complete waste of time that will end up getting me nowhere (and probably worse off than Option 1 where I will actually have a job).

Any help would be greatly appreciated, especially from people working in these fields that can help me out with if I will get a job after spending so many years studying/uptaking a new career path

Sorry about the Engineering Prefix - I accidentally put Engineering and I don't think there is a way to edit it (I can't delete the post and post it again with a different prefix I think).

 

[wukunlin]

About your issues:
1. From my own experience, getting a STEM job in Australia is, in general, far from easy. Physics jobs are pretty much non-existent (unless you count post-docs). Most graduate recruitment process for industry jobs involve which applicant can talk over the other. The ratio of new graduates and new openings is just really ugly.
2. It's not impossible but you are competing against applicants with Astrophysics degrees or degrees specialized in machine learning. If you can establish some great networks though, the degrees may matter less.
3. PhD students I have encountered generally live on stipends/scholarship and teaching assistant jobs.
4. Better late than never, if it is truly your dream
5. Still need to figure out if this is something you truly want to do. If not, it won't matter how great or how fast this pathway is.
6. If you can financially sustain yourself when studying for something your love, I don't see why not.

Other stuff I want to add:
1. how much do you like machine learning exactly? In terms of pursuing astrophysics, are you using it as a stepping stone? I don't believe it is a good one.
2. I don't know much about postgrad stat courses. Thinking back to my first year math courses, if that was all the maths I learned, my brain would've been blown out by postgrad level physics courses. I don't see how astrophysics can be less rigorous than laser physics (which is what I studied).
3. Basically option 1 will make it easier for you to get an industry job, option 2 will be more restricted to academic

 

[StatGuy2000]

About your issues:
1. From my own experience, getting a STEM job in Australia is, in general, far from easy. Physics jobs are pretty much non-existent (unless you count post-docs). Most graduate recruitment process for industry jobs involve which applicant can talk over the other. The ratio of new graduates and new openings is just really ugly.
2. It's not impossible but you are competing against applicants with Astrophysics degrees or degrees specialized in machine learning. If you can establish some great networks though, the degrees may matter less.
3. PhD students I have encountered generally live on stipends/scholarship and teaching assistant jobs.
4. Better late than never, if it is truly your dream
5. Still need to figure out if this is something you truly want to do. If not, it won't matter how great or how fast this pathway is.
6. If you can financially sustain yourself when studying for something your love, I don't see why not.

Other stuff I want to add:
1. how much do you like machine learning exactly? In terms of pursuing astrophysics, are you using it as a stepping stone? I don't believe it is a good one.
2. I don't know much about postgrad stat courses. Thinking back to my first year math courses, if that was all the maths I learned, my brain would've been blown out by postgrad level physics courses. I don't see how astrophysics can be less rigorous than laser physics (which is what I studied).
3. Basically option 1 will make it easier for you to get an industry job, option 2 will be more restricted to academic

At the risk of hijacking this thread, I have read in many places on the Internet (including an article in the online version of the Economist magazine) that Australia has avoided a recession since 1991, suggesting a continuing trend toward economic growth, and thus employment opportunities. If STEM jobs are not easy to come by, what types of jobs (outside of mining, construction, tourism and retail work) are most readily available in Australia?

 

[Choppy]

I have realized now that I REALLY want to work in Astrophysics/Space Exploration.

I have not taken any Physics courses at Uni, but have done Physics up to year 12.

Before getting too deep into this, these two statements jump out at me and suggest that you might be getting a little ahead of yourself. If you've never taken a physics course at university, how can you be so sure that you're want to pursue astrophysics for a PhD? On top of that your second "option" reads like you're trying to get into a PhD in astrophysics without having done any actual physics. I suspect that if you go down this path, you're going to run into some major roadblocks. if you really want to pursue astrophysics - you need some undergraduate coursework in physics to put you on par with anyone else pursuing advanced studies in that field.

 

[Harshna]

About your issues:
1. From my own experience, getting a STEM job in Australia is, in general, far from easy. Physics jobs are pretty much non-existent (unless you count post-docs). Most graduate recruitment process for industry jobs involve which applicant can talk over the other. The ratio of new graduates and new openings is just really ugly.
2. It's not impossible but you are competing against applicants with Astrophysics degrees or degrees specialized in machine learning. If you can establish some great networks though, the degrees may matter less.
3. PhD students I have encountered generally live on stipends/scholarship and teaching assistant jobs.
4. Better late than never, if it is truly your dream
5. Still need to figure out if this is something you truly want to do. If not, it won't matter how great or how fast this pathway is.
6. If you can financially sustain yourself when studying for something your love, I don't see why not.

Other stuff I want to add:
1. how much do you like machine learning exactly? In terms of pursuing astrophysics, are you using it as a stepping stone? I don't believe it is a good one.
2. I don't know much about postgrad stat courses. Thinking back to my first year math courses, if that was all the maths I learned, my brain would've been blown out by postgrad level physics courses. I don't see how astrophysics can be less rigorous than laser physics (which is what I studied).
3. Basically option 1 will make it easier for you to get an industry job, option 2 will be more restricted to academic

At the risk of hijacking this thread, I have read in many places on the Internet (including an article in the online version of the Economist magazine) that Australia has avoided a recession since 1991, suggesting a continuing trend toward economic growth, and thus employment opportunities. If STEM jobs are not easy to come by, what types of jobs (outside of mining, construction, tourism and retail work) are most readily available in Australia?

Before getting too deep into this, these two statements jump out at me and suggest that you might be getting a little ahead of yourself. If you've never taken a physics course at university, how can you be so sure that you're want to pursue astrophysics for a PhD? On top of that your second "option" reads like you're trying to get into a PhD in astrophysics without having done any actual physics. I suspect that if you go down this path, you're going to run into some major roadblocks. if you really want to pursue astrophysics - you need some undergraduate coursework in physics to put you on par with anyone else pursuing advanced studies in that field.

Hi all,

Thank you so much for your replies - it is really helping and much appreciated. I will reply to everyone separately below

 

[wukunlin]

At the risk of hijacking this thread, I have read in many places on the Internet (including an article in the online version of the Economist magazine) that Australia has avoided a recession since 1991, suggesting a continuing trend toward economic growth, and thus employment opportunities. If STEM jobs are not easy to come by, what types of jobs (outside of mining, construction, tourism and retail work) are most readily available in Australia?

Happy to move this to a private conversation if necessary...
I wasn't aware of readily available jobs when I was job seeking as a fresh gradate in 2014. I spent around 6 months looking for jobs with a half-physics half-engineering degree, only got a couple of offers to work in one of those big 4 accounting firms. The job market is probably different for experienced professionals I think.

 

[Harshna]

2. It's not impossible but you are competing against applicants with Astrophysics degrees or degrees specialized in machine learning. If you can establish some great networks though, the degrees may matter less.

Hi wukunlin,
Thank you so much for addressing all my issues - I really appreciate it and they are useful! Sorry I should have added in my original post that what I meant was after gaining some experience in Machine Learning than becoming more of an expert than maybe I could get a Physics Research Institute, etc wanting to employ me? Not sure if this is a real expectation though :)

Other stuff I want to add:
1. how much do you like machine learning exactly? In terms of pursuing astrophysics, are you using it as a stepping stone? I don't believe it is a good one.
2. I don't know much about postgrad stat courses. Thinking back to my first year math courses, if that was all the maths I learned, my brain would've been blown out by postgrad level physics courses. I don't see how astrophysics can be less rigorous than laser physics (which is what I studied).
3. Basically option 1 will make it easier for you to get an industry job, option 2 will be more restricted to academic

1. I like Machine Learning as it involves maths and I find it quite interesting how we can predict things using machine learning. I'm kind of thinking of applying Astrophysics/Space Exploration to Machine Learning like predicting what will happen in astrophysics problems using machine learning
2. I think my Uni is actually a bit different with this in that they will account for people who have not done any courses in that field before by teaaching the basics first - so these specific postgrad courses have no prerequisites. My Graduate Diploma of Science was also all postgrad courses and I had never done a course in Data Science ever I think the Astronomy courses I would be doing in my Master in Applied Data Science - their course descriptions and topics have no mention of anything mathematical - just theory (which disappointed me too because I like maths), so I guess it would be easier without doing any undergrad courses in it I guess? The links for these courses:

https://www.usq.edu.au/course/synopses/2017/PHY8001.html
https://www.usq.edu.au/course/synopses/2017/PHY8002.html
https://www.usq.edu.au/course/synopses/2017/PHY8003.html
https://www.usq.edu.au/course/synopses/2017/PHY8004.html

3. I totally agree - I am actually planning to get into research (academic)

 

[Harshna]

At the risk of hijacking this thread, I have read in many places on the Internet (including an article in the online version of the Economist magazine) that Australia has avoided a recession since 1991, suggesting a continuing trend toward economic growth, and thus employment opportunities. If STEM jobs are not easy to come by, what types of jobs (outside of mining, construction, tourism and retail work) are most readily available in Australia?

From my experiences I think the STEM industry is new and just starting to grow. I have found that I think data science, computer science and IT can support you to get a job in STEM. I have personally found that doing these can help you become something in STEM fields such as Analyst of Health Data (there are plenty of Analyst jobs out there). I have myself worked as an Analyst in Insurance (but science has always been my passion so I was not happy with this role). I am not sure about Physics though and can't really find anything even combining with data science, computer science and IT - but I am only new to this field and hence I myself are trying to explore options because I am too very unsure and have big decisions to make right now - so am myself looking for answers

Hope this helps

 

[Harshna]

Before getting too deep into this, these two statements jump out at me and suggest that you might be getting a little ahead of yourself. If you've never taken a physics course at university, how can you be so sure that you're want to pursue astrophysics for a PhD? On top of that your second "option" reads like you're trying to get into a PhD in astrophysics without having done any actual physics. I suspect that if you go down this path, you're going to run into some major roadblocks. if you really want to pursue astrophysics - you need some undergraduate coursework in physics to put you on par with anyone else pursuing advanced studies in that field.

Hi Choppy,

Thank you so much for your reply - it is useful!

I have always like physics and maths and have been reading and thinking a lot about astrophysics and space exploration and it has just fascinated me so much

I think my Uni is actually a bit different with this in that they will account for people who have not done any courses in that field before by teaaching the basics first - so these specific postgrad courses have no prerequisites. My Graduate Diploma of Science was also all postgrad courses and I had never done a course in Data Science ever I think the Astronomy courses I would be doing in my Master in Applied Data Science - their course descriptions and topics have no mention of anything mathematical - just theory (which disappointed me too because I like maths), so I guess it would be easier without doing any undergrad courses in it I guess? The links for these courses:
https://www.usq.edu.au/course/synopses/2017/PHY8001.html
https://www.usq.edu.au/course/synopses/2017/PHY8002.html
https://www.usq.edu.au/course/synopses/2017/PHY8003.html
https://www.usq.edu.au/course/synopses/2017/PHY8004.htm

My plan is to do the Masters and then Masters by Research and then PhD. I agree it could be a bit of an overshot but for some reason I feel confident to do these courses lol

 

[wukunlin]

1. I think my Uni is actually a bit different with this in that they will account for people who have not done any courses in that field before by teaaching the basics first - so these specific postgrad courses have no prerequisites. My Graduate Diploma of Science was also all postgrad courses and I had never done a course in Data Science ever I think the Astronomy courses I would be doing in my Master in Applied Data Science - their course descriptions and topics have no mention of anything mathematical - just theory (which disappointed me too because I like maths), so I guess it would be easier without doing any undergrad courses in it I guess? The links for these courses:

https://www.usq.edu.au/course/synopses/2017/PHY8001.html
https://www.usq.edu.au/course/synopses/2017/PHY8002.html
https://www.usq.edu.au/course/synopses/2017/PHY8003.html
https://www.usq.edu.au/course/synopses/2017/PHY8004.html

That's... a bit surprising. Although, I don't think there is a way to substitute a 4 year undergrad degree in physics without spending just as much time on it. There are a lot of fundamental concepts that takes a couple of years to develop. This is something your potential PhD supervisor will most likely consider. Having said that, if said supervisor desperately need someone to do machine learning for his or her astrophysics application, an expert in machine learning who understands the astrophysics research goal will have an advantage.

 

[Harshna]

That's... a bit surprising. Although, I don't think there is a way to substitute a 4 year undergrad degree in physics without spending just as much time on it. There are a lot of fundamental concepts that takes a couple of years to develop. This is something your potential PhD supervisor will most likely consider. Having said that, if said supervisor desperately need someone to do machine learning for his or her astrophysics application, an expert in machine learning who understands the astrophysics research goal will have an advantage.

Thanks so much for your reply. This is along the thoughts of what I was thinking - I think machine learning is going to just keep getting bigger and machine learning is already probably one of the most sought out skills in any industry and I think with machine learning and some knowledge of another field (astrophysics) you should be able to get a job in it, I guess? Glad to know that I am not with the fairies on this and someone else that knows the physics industry is also thinking along the same lines!

I agree it is better to do an undergraduate degree, but I'm thinking it could be too late for me - it already is taking 6-7 years without the undergraduate degree, so looking for the fastest option to do it if it will work out

 

[CrysPhys]

From my experiences I think the STEM industry is new and just starting to grow. I have found that I think data science, computer science and IT can support you to get a job in STEM. I have personally found that doing these can help you become something in STEM fields such as Analyst of Health Data (there are plenty of Analyst jobs out there). I have myself worked as an Analyst in Insurance (but science has always been my passion so I was not happy with this role). I am not sure about Physics though and can't really find anything even combining with data science, computer science and IT - but I am only new to this field and hence I myself are trying to explore options because I am too very unsure and have big decisions to make right now - so am myself looking for answers

Hope this helps

You might be interested in this government report on the Australian STEM workforce:

http://www.chiefscientist.gov.au/2016/03/report-australias-stem-workforce/

The report was issued in 2016, but based on a 2011 census. A new report will be prepared in late 2018, based on the 2016 census, according to the report.

 

[Harshna]

You might be interested in this government report on the Australian STEM workforce:

http://www.chiefscientist.gov.au/2016/03/report-australias-stem-workforce/

The report was issued in 2016, but based on a 2011 census. A new report will be prepared in late 2018, based on the 2016 census, according to the report.

Wow - thank you for this - that is interesting! Are there any more of these reports somewhere - this would be interesting to learn as much out there as possible on jobs like specific fields, etc

 

[Choppy]

I can't say that I'm all that familiar with the Australian system. But if I'm understanding things, there's something that's not adding up for me here.

think the Astronomy courses I would be doing in my Master in Applied Data Science - their course descriptions and topics have no mention of anything mathematical - just theory (which disappointed me too because I like maths), so I guess it would be easier without doing any undergrad courses in it I guess?

Are these graduate-level courses? If so, it seems highly unlikely that there would be no prerequisites for them. And if there are not, or if they are actually undergraduate level survey courses that don't get into the math, then they're not preparing you for graduate studies in physics.

On top of that you're talking about taking them while completing a master's degree in applied data science, part-time, that will only take you one year to complete. When are you going to learn the applied data science in the applied data science master's degree?

I think with machine learning and some knowledge of another field (astrophysics) you should be able to get a job in it, I guess? Glad to know that I am not with the fairies on this and someone else that knows the physics industry is also thinking along the same lines!

An astrophysics PhD supervisor who has a project that involves machine learning is not going to seriously consider taking on a student who knows machine learning and took a handful of courses that didn't involve any of the rigorous mathematical background required to do astrophysics. This would set the student up for failure. How is such a student going to pass qualifying exams? A candidacy exam? Complete advanced coursework in core subjects like quantum mechanics or electrodynamics?

There isn't really an "astrophysics industry." There is academia, and then the aerospace industry. With a background in machine learning, you might be able to get a job in the aerospace industry. But academia is highly competitive and saturated with people who have PhDs and many of them also either know, or can learn the necessary machine learning that they need.

 

[Rika]

You should ask yourself not if you should do astrophysics phd because that's your dream but if you fail to get a job in astrophysics afterwards - will you regret it? Because it's almost impossible to get a job in astrophysics nowadays so if you choose option no 2 you should think about it as a fun adventure rather than something that can boost your career.

 

[Harshna]

You should ask yourself not if you should do astrophysics phd because that's your dream but if you fail to get a job in astrophysics afterwards - will you regret it? Because it's almost impossible to get a job in astrophysics nowadays so if you choose option no 2 you should think about it as a fun adventure rather than something that can boost your career.

Hi Rika,
Thank you so much for the insight and help - all these replies are so helpful to help me in deciding and the more replies I can get for the most clear perspective possible the better, so it is much appreciated

 

[Harshna]

I can't say that I'm all that familiar with the Australian system. But if I'm understanding things, there's something that's not adding up for me here.
Are these graduate-level courses? If so, it seems highly unlikely that there would be no prerequisites for them. And if there are not, or if they are actually undergraduate level survey courses that don't get into the math, then they're not preparing you for graduate studies in physics.

On top of that you're talking about taking them while completing a master's degree in applied data science, part-time, that will only take you one year to complete. When are you going to learn the applied data science in the applied data science master's degree?An astrophysics PhD supervisor who has a project that involves machine learning is not going to seriously consider taking on a student who knows machine learning and took a handful of courses that didn't involve any of the rigorous mathematical background required to do astrophysics. This would set the student up for failure. How is such a student going to pass qualifying exams? A candidacy exam? Complete advanced coursework in core subjects like quantum mechanics or electrodynamics?

There isn't really an "astrophysics industry." There is academia, and then the aerospace industry. With a background in machine learning, you might be able to get a job in the aerospace industry. But academia is highly competitive and saturated with people who have PhDs and many of them also either know, or can learn the necessary machine learning that they need.

Hi Choppy,

Sorry - I missed your reply. These are postgraduate level courses as part of the Masters.

I have done a Graduate Diploma in Applied Data Science so I have done most of the Data Science courses (except for one). I would only have two research/industry courses (2 units each and depending if I can get a workplace, otherwise we have to take the research courses instead) and I would need two electives (if I credit other courses I have done), but there is space for 4 electives if I don't apply for credit. So I have 5 courses (7 units to do with the 2 unit courses) and I have already done one of those courses this semester - so 1 year part time Sorry I should have clarified that earlier - I am fairly new on Physics Forums

I agree - but I'm kind of thinking along the track of how many people would actually have experience in both machine learning and another field, so I am guessing they won't expect both? Actually the research scholarship I have was offered to me by my Supervisor who had no idea that I did Biology (I did my Biology Degree at a different Uni to this one), but he knew that I did Data Science - this is what is making me wonder... but still not entirely sure how much this reflects to the real working world - so I really appreciate your perspective on this from a real industry professional which is helping me clear my head and know what really the situation out there is

Interesting that there isn't an astrophysics industry and it does make sense that there would be a lot of people who know it in academia and can learn it - some more food for thought for me

 

[quantknight]

I can't say that I'm all that familiar with the Australian system. But if I'm understanding things, there's something that's not adding up for me here.
Are these graduate-level courses? If so, it seems highly unlikely that there would be no prerequisites for them. And if there are not, or if they are actually undergraduate level survey courses that don't get into the math, then they're not preparing you for graduate studies in physics.

On top of that you're talking about taking them while completing a master's degree in applied data science, part-time, that will only take you one year to complete. When are you going to learn the applied data science in the applied data science master's degree?An astrophysics PhD supervisor who has a project that involves machine learning is not going to seriously consider taking on a student who knows machine learning and took a handful of courses that didn't involve any of the rigorous mathematical background required to do astrophysics. This would set the student up for failure. How is such a student going to pass qualifying exams? A candidacy exam? Complete advanced coursework in core subjects like quantum mechanics or electrodynamics?

There isn't really an "astrophysics industry." There is academia, and then the aerospace industry. With a background in machine learning, you might be able to get a job in the aerospace industry. But academia is highly competitive and saturated with people who have PhDs and many of them also either know, or can learn the necessary machine learning that they need.

Hi choppy, you mentioned there are jobs in aerospace for machine learning with physics, would you please explain what sort of jobs are they? And also what other physics branches that can be combined with machine learning which will have a industry jobs??

 

[Choppy]

Hi choppy, you mentioned there are jobs in aerospace for machine learning with physics, would you please explain what sort of jobs are they? And also what other physics branches that can be combined with machine learning which will have a industry jobs??

Well, what I said was that with a background in machine learning the OP might be able to get a job in the aerospace industry. That is admittedly a supposition on my part as I'm not intimately familiar with the field, but like any engineering-heavy industry, there is likely to be at least some demand for machine learning in the near future.

As an example I could imagine a new airplane being designed to interface with a neural network that monitors various properties of its important components: operational temperatures, pressures, age, etc. and identifying likely failures well before the component fails and even ordering replacements and scheduling preventative maintenance. We're seeing this happen right now in the medical industry and I don't think we're unique.

In medical physics machine learning is becoming popular for tasks such as computer-assisted diagnoses. Computers can process a lot more information from a medical image than a human can and so the game of "puppy-or-muffin" can easily turn into "tumour-not-a-tumour." But on top of that there are things like workflow and process control, automated treatment planning, treatment optimization, and medical decision making that can be assisted by intelligent machines.

Any branch of physics that involves processing large amounts of data is likely to benefit from machine learning. Think about astronomy and the search for extra-solar planets, or geophysics and the search for oil.