Regarding (undergraduate) honors thesis

[TenaciousBushLeaper]

Hi all, 

I'm currently in the second semester of my senior honors thesis. To give a very, very brief description of my work, it's on word reading. For the first part of the project, I collected, read, and sorted over 12,000 unique responses from participants in the form of definitions for words.  I did most of this over the course of the last semester (Fall 2014) and am just about done, only have 900ish left. Now, I'm a bit burnt out at this point. & I have to get on with the project or else I might not be able to complete it before graduation.  (Oh, and I even worked on this just about everyday over winter break). The second part of the project isn't nearly as much work in terms of data collection but I still have to write up the manuscript. 

I'm wondering if this actually isn't all that much work and I'm just for the lack of a better description, complaining and being lazy?  What do you think? 

Any suggestions on how crank up my motivation ? Also I should mention last semester I enrolled in 20 credit hours. This semester I cut it down to 15 for the sake of my sanity. 

I'd also like to hear about the type of projects other undergraduates complete. 

Edited February 1, 2015 by TenaciousBushLeaper

[TXInstrument11]

I have a few free response questions in a survey I am using, though obviously not to as great an extent as you have. Is there possibly any way you can group them by keywords or  something? Excel should be able to do this, as could Qualtrics, if you are using that software for your surveys. 

 

Also, what is your N? I figure the 12K refers to each individual question/definition, no? If you are in the hundreds for sample size, you could perhaps just cut it  off and go straight to data analysis. If you are concerned about the second half of your responses involving some kind of bias, you could run preliminary tests for average response length and keywords to get an idea of how similar the rest of the sample is to your completed group. Then, if they look pretty darn similar, you could have a somewhat decent case for ignoring these other responses. 

 

My partially free-response survey is the Ambiguous Intentions Hostility Questionnaire (AIHQ). Thankfully, the free response portions are not necessary to derive an overall score for it. I am likely going to sort and color code with keywords to help cut down on my work load and/or recruit some other undergraduates in the lab to help me sort them. I may also run some analyses on response length and see if that indicates anything strange or interesting. 

[TenaciousBushLeaper]

My sample size is 200, 200 participants gave up to 5 different definitions for different sets of words, each set containing 60 words, with 25 participants working on a particular set. So 8 sets, 25 participants per set & 60 words per set. At the outset I attempted creating a natural language parser using python, so that my work would be cut by some & succeeded but the results were sub par and unreliable.  I can't extrapolate on a portion of the data since each definition is specific to a word and the content of the definition is the valuable information. 

Also, the platform on which participants provided their response was not any type of proprietary software package/website. The website and the task were individually coded and set up on a server on my university. 

[Tacoma]

Hi all, 

I'm currently in the second semester of my senior honors thesis. To give a very, very brief description of my work, it's on word reading. For the first part of the project, I collected, read, and sorted over 12,000 unique responses from participants in the form of definitions for words.  I did most of this over the course of the last semester (Fall 2014) and am just about done, only have 900ish left. Now, I'm a bit burnt out at this point. & I have to get on with the project or else I might not be able to complete it before graduation.  (Oh, and I even worked on this just about everyday over winter break). The second part of the project isn't nearly as much work in terms of data collection but I still have to write up the manuscript. 

I'm wondering if this actually isn't all that much work and I'm just for the lack of a better description, complaining and being lazy?  What do you think? 

Any suggestions on how crank up my motivation ? Also I should mention last semester I enrolled in 20 credit hours. This semester I cut it down to 15 for the sake of my sanity. 

I'd also like to hear about the type of projects other undergraduates complete. 

 

I don't think you're being lazy at all. An honours thesis is easily as much work as an average class if not more but with far less structure and guidance. Although for me at least I have found that the very beginning of the year and the very end are where most of the hard work takes place/will take place (i.e., ethics application and revisions at the beginning and data analysis/manuscript preparation at the end). 

 

Also, are the courses at your university worth 5 credits each? Just wondering how many courses 20 credit hours works out to? Where I go to school that would be 6 or 7 classes as they are 3 credits each and I've seen other schools where classes are 1.5 credits each. 

[TXInstrument11]

My sample size is 200, 200 participants gave up to 5 different definitions for different sets of words, each set containing 60 words, with 25 participants working on a particular set. So 8 sets, 25 participants per set & 60 words per set. At the outset I attempted creating a natural language parser using python, so that my work would be cut by some & succeeded but the results were sub par and unreliable.  I can't extrapolate on a portion of the data since each definition is specific to a word and the content of the definition is the valuable information. 

Also, the platform on which participants provided their response was not any type of proprietary software package/website. The website and the task were individually coded and set up on a server on my university. 

Wow, that's intense. I'm sorry, but I have no advice to offer you on how to go about fixing this, especially if you've already gone about trying to automate the process through programming. Do your advisors/thesis members have any advice on this?

 

Do any of  your sources have similar designs? I've emailed professors before about how to go about scoring and interpreting surveys we've both used. Maybe someone else has created a decent method to cut down on the work required here. Since you're an undergrad, they may take pity. Alternately, they may actually be interested in your results! A professor I just emailed for access to his survey scoring materials expressed that he would be interested in learning the results of my experiment and gladly handed over the information I needed.

[TenaciousBushLeaper]

I don't think you're being lazy at all. An honours thesis is easily as much work as an average class if not more but with far less structure and guidance. Although for me at least I have found that the very beginning of the year and the very end are where most of the hard work takes place/will take place (i.e., ethics application and revisions at the beginning and data analysis/manuscript preparation at the end). 

 

Also, are the courses at your university worth 5 credits each? Just wondering how many courses 20 credit hours works out to? Where I go to school that would be 6 or 7 classes as they are 3 credits each and I've seen other schools where classes are 1.5 credits each. 

The courses at my university are the standard (I assume) 3 - 4 credit courses. I had two 4 credit courses and four 3 credit courses.  

 

 

Wow, that's intense. I'm sorry, but I have no advice to offer you on how to go about fixing this, especially if you've already gone about trying to automate the process through programming. Do your advisors/thesis members have any advice on this?

 

Do any of  your sources have similar designs? I've emailed professors before about how to go about scoring and interpreting surveys we've both used. Maybe someone else has created a decent method to cut down on the work required here. Since you're an undergrad, they may take pity. Alternately, they may actually be interested in your results! A professor I just emailed for access to his survey scoring materials expressed that he would be interested in learning the results of my experiment and gladly handed over the information I needed.

My advisor's solution was to hire help, in the form of an RA working for 3 credits. In short, the help quit after finishing 1 set (each set contained approx 1500 responses) & failed to send me her completed work. 

Usually the type of work I'm doing which is, partially, creating a resource for future research, is done on a larger time frame with more than one person working on it. My advisor has told me I've done a significant amount of work & that a write up of the work I've done so far would be sufficient however, there would be no experiment. I'd have simply created a resource & this alone doesn't sit well with me. 

Edited February 2, 2015 by TenaciousBushLeaper

[Tacoma]

The courses at my university are the standard (I assume) 3 - 4 credit courses. I had two 4 credit courses and four 3 credit courses.  

 

 

Ya, if you're doing six courses at a time including your honours thesis you're definitely not lazy. I only have three classes right now while I'm doing my honours plus TA and RA work and that is more than enough to keep me busy so good on you for being able to handle six. 

[TenaciousBushLeaper]

Ya, if you're doing six courses at a time including your honours thesis you're definitely not lazy. I only have three classes right now while I'm doing my honours plus TA and RA work and that is more than enough to keep me busy so good on you for being able to handle six. 

Well, I was enrolled in six the previous semester (first half of the senior thesis). This semester I am enrolled in 5, each 3 credits (including senior thesis).

Could you tell me about your thesis? 

Edited February 2, 2015 by TenaciousBushLeaper

[Tacoma]

Well, I was enrolled in six the previous semester (first half of the senior thesis). This semester I am enrolled in 5, each 3 credits (including senior thesis).

Could you tell me about your thesis? 

 

Ya, six classes would burn anyone out so no wonder you felt that way. My honours thesis examines prejudice in the judicial system as well as the impact of computer mediated communication (CMC) versus face-to-face (Ftf) communication.

 

Participants come to my study and read one of two criminal history reports. The history report states that a perpetrator was found guilty of severely injuring (gunshot wound followed by lengthy hospital stay) a homeowner while breaking into a home and that they (the participant) must now suggest a prison sentence for perpetrator in years and months up to a maximum of 25 years. The two criminal history reports are identical in every way excepting for the race of the perpetrator, which is manipulated as being either Caucasian or Aboriginal. After reading the report, the participants are asked for their suggested sentence as well as a few follow up questions before filling out the Marlowe-Crowne social desirability scale and a demographics questionnaire.

 

The other manipulation is whether the participant is in the CMC or Ftf condition. In the CMC condition participants provide their sentence and answer the follow up questions over a lab computer while in the Ftf condition the participants give their suggested sentence and answer the follow up questions in person. The research questions I am interested in are: 1). Do people assign more punitive sentences to Aboriginals than to Caucasians; and 2.) Do people assign more punitive sentences in CMC than in Ftf situations.  

[smbtuckma]

Ugh man I tried Python NLP for a project a few years ago. Frustrating, but maybe in 10 years or so it'll be reliable...

You're not lazy at all, and I'm super impressed you're only now starting to burn out. I start data collection next week, and then have to kick it into high gear in terms of getting our analysis software to work since right now the new version of matlab screwed it up. Im worried about it all building up at the end because of problems there. And even though I only have that plus two courses (thesis counts as a full course here), I also have four music ensembles I'm dedicated to and a TA position. On paper the schedule may look easier, but this stuff takes just as much time as a full course load haha

Edited February 2, 2015 by smbtuckma

[Chubberubber]

My advisor has told me I've done a significant amount of work & that a write up of the work I've done so far would be sufficient however, there would be no experiment. I'd have simply created a resource & this alone doesn't sit well with me. 

 

Seems like you're in a good place in terms of progress, especially since your advisor told you you've done enough. There's a time and a place to overachieve, you just need to decide whether now is right time for you to do it. If it is- use the fact that you're doing more than you have to as your motivation! If it isn't- remember that researchers often work on one "Mega Experiment" like yours for years and publish bits and pieces of it at a time. The full scale experiment you're planning can be the basis of a future PhD dissertation, so you don't have to feel compelled to sinish it all right now...