Professional help writing a PhD dissertation

[Moshe Marko]

Hello,

I am a PhD student living in the US. English is my second language. Unfortunately I had a very limited help with the design and the writing of the dissertation paper. A week ago I successfully defended my PhD dissertation but I was asked to make some major editorial changes. I was also asked to improve my graphs and tables. I have to say that for the last few days I have been working around the clock to try and improve the paper but I find this task to be overwhelmingly difficult for me to do by myself.  Does anyone know any reliable professional person or agency that can help me edit the paper and improve the visual representation of the data?

Thanks

Moshe

Moshe Marko, PT, DPT, MHS, OCS, CSCS
Assistant Professor
Department of Physical Therapy Education
College of Health Professions
SUNY Upstate Medical University
Room 2232 Silverman Hall
750 Adams Street
Syracuse, NY 13210-1834
315 464 6577
FAX [phone]
[hidden email]

[Art Kendall]

I do not know anybody in your area.

However, you might do better by splitting your request.  You need a person skilled at text editing which you may be able to find in your English or Journalism department.

You need a person with SPSS table presentation and graphics skills to help you polish your tables and graphs. You might find this person in any of the social science departments.

Also, you might get a lot of the English polished if you have WordPerfect.  It has great style checking.  If you do not, the checking in Word is not too bad.  Beware that it often fails to detect "street English".

Art Kendall
Social Research Consultants



On 9/27/2011 11:11 PM, Moshe Marko wrote:

Hello,

I am a PhD student living in the US. English is my second language. Unfortunately I had a very limited help with the design and the writing of the dissertation paper. A week ago I successfully defended my PhD dissertation but I was asked to make some major editorial changes. I was also asked to improve my graphs and tables. I have to say that for the last few days I have been working around the clock to try and improve the paper but I find this task to be overwhelmingly difficult for me to do by myself.  Does anyone know any reliable professional person or agency that can help me edit the paper and improve the visual representation of the data?

Thanks

Moshe

Moshe Marko, PT, DPT, MHS, OCS, CSCS
Assistant Professor
Department of Physical Therapy Education
College of Health Professions
SUNY Upstate Medical University
Room 2232 Silverman Hall
750 Adams Street
Syracuse, NY 13210-1834
315 464 6577
FAX [phone]
[hidden email]

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Sonia Brandon-2]

I've used this service before - they were quite helpful.

 

http://edit911.com/

>>> Art Kendall <[hidden email]> 9/29/2011 8:38 AM >>>
I do not know anybody in your area.

However, you might do better by splitting your request.  You need a person skilled at text editing which you may be able to find in your English or Journalism department.

You need a person with SPSS table presentation and graphics skills to help you polish your tables and graphs. You might find this person in any of the social science departments.

Also, you might get a lot of the English polished if you have WordPerfect.  It has great style checking.  If you do not, the checking in Word is not too bad.  Beware that it often fails to detect "street English".

Art Kendall
Social Research Consultants



On 9/27/2011 11:11 PM, Moshe Marko wrote:

Hello,

I am a PhD student living in the US. English is my second language. Unfortunately I had a very limited help with the design and the writing of the dissertation paper. A week ago I successfully defended my PhD dissertation but I was asked to make some major editorial changes. I was also asked to improve my graphs and tables. I have to say that for the last few days I have been working around the clock to try and improve the paper but I find this task to be overwhelmingly difficult for me to do by myself.  Does anyone know any reliable professional person or agency that can help me edit the paper and improve the visual representation of the data?

Thanks

Moshe

Moshe Marko, PT, DPT, MHS, OCS, CSCS
Assistant Professor
Department of Physical Therapy Education
College of Health Professions
SUNY Upstate Medical University
Room 2232 Silverman Hall
750 Adams Street
Syracuse, NY 13210-1834
315 464 6577
FAX [phone]
[hidden email]

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[John F Hall]

Not sure what your data look like or what your university requirements are, but you might find something useful on report writing (especially graphics) in:

 

http://www.audiencedialogue.net/kya6b.html

 

 

John F Hall

 

[hidden email]

www.surveyresearch.weebly.com

 

 

 

[John F Hall]

Stace

 

I’m not a statistician and can’t really help you on this, so am copying this reply to the list where you will get better advice.  However, you need to give much more detail.  Also, have you thought about the SRMS section of the ASA?

 

 

John F Hall

 

[hidden email]

www.surveyresearch.weebly.com

 

 

 

 

From: stace swayne [mailto:[hidden email]]
Sent: 04 October 2011 15:08
To: John F Hall
Subject: Some advice on SPSS sample selections

 

Dear John,

 

 I need some guidance on what type of syntax I need in order to do the following:

In my study there will be 400 subjects, 200 in a control group and 200 in a treatment group (people who will receive specialized job training).
 

My colleagues have recruited people who fall into three categories (1)  recently unemployed six months or less, (2) unemployed for 12 months and  (3) unemployed for greater than 12 months.

Members of my group are interested in having different proportions of the  people who fall into the aforementioned categories represented in the treatment group. For example, 20 percent should come from the recently
 unemployed or 30 percent from the unemployed for greater than 12 months  category etc.

Can you give me some advice on how to do this via syntax?

Thank you,

 

Stace

[Art Kendall]

Please give more detail. for example:

Have the subjects already been selected and you want to randomly assign them to treatment group being sure that each employment stratum is evenly divided?

Or are you going to draw a sample from each of three known populations? Or are you going to draw a large sample and post-stratify?

How were the percentages in the strata chosen?  Proportional to stratum pop?  "over sampling" small strata to facilitate comparisons?

How did you decide on an N of 400?



Art Kendall
Social Research Consultants

On 10/4/2011 2:58 PM, John F Hall wrote:

Stace

 

I’m not a statistician and can’t really help you on this, so am copying this reply to the list where you will get better advice.  However, you need to give much more detail.  Also, have you thought about the SRMS section of the ASA?

 

 

John F Hall

 

[hidden email]

www.surveyresearch.weebly.com

 

 

 

 

From: stace swayne [[hidden email]]
Sent: 04 October 2011 15:08
To: John F Hall
Subject: Some advice on SPSS sample selections

 

Dear John,

 

 I need some guidance on what type of syntax I need in order to do the following:

In my study there will be 400 subjects, 200 in a control group and 200 in a treatment group (people who will receive specialized job training).
 

My colleagues have recruited people who fall into three categories (1)  recently unemployed six months or less, (2) unemployed for 12 months and  (3) unemployed for greater than 12 months.

Members of my group are interested in having different proportions of the  people who fall into the aforementioned categories represented in the treatment group. For example, 20 percent should come from the recently
 unemployed or 30 percent from the unemployed for greater than 12 months  category etc.

Can you give me some advice on how to do this via syntax?

Thank you,

 

Stace

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Art Kendall]

Members of my group are interested in having different proportions of the� � people who fall into the aforementioned categories represented in the treatment group.

Deciding relative stratum sampling size for strata on a basis other than 1) proportional to representation in order to have efficient estimation of totals, mean, pop percents etc.,� or 2) equal for statistical efficiency, raises a red flag.� � I strongly suggest that you work out with your client what the purposes of the research are.� What questions are the team trying to answer?

Since efficiency in comparing treatment and control groups� is most likely of major interest, within each stratum you would have equal numbers of treatment and control cases.

The power analysis gives you an estimate of what size achieved sample you have.� If there are pre-tests, how long is the time interval between any pre-tests and the last post-test? What kind of non-response and attrition do you anticipate?� You need to draw more cases so that you end up with the sample sizes from the power analysis.

Is the middle stratum actually 6 to 12 months?� What is the rationale for stratifying (coarsening your measure of time unemployed) rather than sing time as a continuous variable?

Will recruiting continue until you reach all three stratum quotas?
Will the treatment be given individually or to the entire treated part of the stratum.

something like this UNTESTED syntax would be a place to start.

numeric cut_date(adate10).
compute cut_date = date.mdy(11,21,2011).
compute unemployed_months = datediff(cut_date, date_unemployed, "months").
recode unemployed_months (lo thru 6 =1)(6 thru 12=2)(12 thru high =3) (else=-1) into unemployed_stratum.
missing values unemployed_stratum (-1).
value labels unemployed_stratum
� � � -1 'could not calculate'
� � � 1 'up to 6 months'
� � � 2 'between 6 and 12 months'
� � � 3 'over 12 months'.
Do if unemployed_stratum 1.
� � � sample 120 from 320.
� � � else if unemployed_stratum eq 2.
� � � sample 135 from 615.
� � � else if unemployed_stratum eq 3.
� � � sample 145 from 900.
� � � else.
� � � Print /"oops! problem getting unemployment stratum", ID.
end if.
sort cases by unemployed_stratum.
compute ranorder = rv.uniform(0,2e31).
compute treat_group = mod($casenum,2).
value labels treat_group
� � � 0 'control'
� � � 1 'treatment'.

Art Kendall
Social Research Consultants


On 10/5/2011 9:37 AM, stace swayne wrote:

Hi Art,

�

Below is a response to your questions:

�

Have the subjects already been selected and you want to randomly assign them to treatment group being sure that each employment stratum is evenly divided?

The subjects have not been selected yet we are in the recruitment stage of the project.

�

Or are you going to draw a sample from each of three known populations? Or are you going to draw a large sample and post-stratify?

The goal is to recruit a large sample and then post-stratify.

How were the percentages in the strata chosen?� Proportional to stratum pop?� "over sampling" small strata to facilitate comparisons?

The percentages were chosen based on the interests of the community based organizations the primary investigator is working with. This is a small pilot community based study.

How did you decide on an N of 400?

The N of 400 was based on a power analysis as well as cost.

�

From: Art Kendall [hidden email]
To: [hidden email]
Sent: Tuesday, October 4, 2011 3:29 PM
Subject: Re: Some advice on SPSS sample selections

Please give more detail. for example:

Have the subjects already been selected and you want to randomly assign them to treatment group being sure that each employment stratum is evenly divided?

Or are you going to draw a sample from each of three known populations? Or are you going to draw a large sample and post-stratify?

How were the percentages in the strata chosen?� Proportional to stratum pop?� "over sampling" small strata to facilitate comparisons?

How did you decide on an N of 400?



Art Kendall
Social Research Consultants

On 10/4/2011 2:58 PM, John F Hall wrote:

Stace

�

I’m not a statistician and can’t really help you on this, so am copying this reply to the list where you will get better advice.� However, you need to give much more detail.� Also, have you thought about the SRMS section of the ASA?

�

�

John F Hall

�

[hidden email]

www.surveyresearch.weebly.com

�

�

�

�

From: stace swayne [[hidden email]]
Sent: 04 October 2011 15:08
To: John F Hall
Subject: Some advice on SPSS sample selections

�

Dear John,

�

� I need some guidance� on what type of syntax I need in order to do the following:

In my study there will be 400 subjects, 200 in a control group and 200 in a treatment group (people who will receive specialized job training).
�

My colleagues have recruited people who fall into three categories (1)� � recently unemployed six months or less, (2) unemployed for 12 months and� � (3) unemployed for greater than 12 months.

Members of my group are interested in having different proportions of the� � people who fall into the aforementioned categories represented in the treatment group. For example, 20 percent should come from the recently
� unemployed or 30 percent from the unemployed for greater than 12 months� � category etc.

Can� you give me some advice on how to do this via syntax?

Thank you,

�

Stace

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Art Kendall]

note that in my example syntax I did not inflate the requested ample sizes for non-response and attrition.

Art Kendall
Social Research Consultants


On 10/5/2011 11:40 AM, Art Kendall wrote:

Members of my group are interested in having different proportions of the� � people who fall into the aforementioned categories represented in the treatment group.

Deciding relative stratum sampling size for strata on a basis other than 1) proportional to representation in order to have efficient estimation of totals, mean, pop percents etc.,� or 2) equal for statistical efficiency, raises a red flag.� � I strongly suggest that you work out with your client what the purposes of the research are.� What questions are the team trying to answer?

Since efficiency in comparing treatment and control groups� is most likely of major interest, within each stratum you would have equal numbers of treatment and control cases.

The power analysis gives you an estimate of what size achieved sample you have.� If there are pre-tests, how long is the time interval between any pre-tests and the last post-test? What kind of non-response and attrition do you anticipate?� You need to draw more cases so that you end up with the sample sizes from the power analysis.

Is the middle stratum actually 6 to 12 months?� What is the rationale for stratifying (coarsening your measure of time unemployed) rather than sing time as a continuous variable?

Will recruiting continue until you reach all three stratum quotas?
Will the treatment be given individually or to the entire treated part of the stratum.

something like this UNTESTED syntax would be a place to start.

numeric cut_date(adate10).
compute cut_date = date.mdy(11,21,2011).
compute unemployed_months = datediff(cut_date, date_unemployed, "months").
recode unemployed_months (lo thru 6 =1)(6 thru 12=2)(12 thru high =3) (else=-1) into unemployed_stratum.
missing values unemployed_stratum (-1).
value labels unemployed_stratum
� � � -1 'could not calculate'
� � � 1 'up to 6 months'
� � � 2 'between 6 and 12 months'
� � � 3 'over 12 months'.
Do if unemployed_stratum 1.
� � � sample 120 from 320.
� � � else if unemployed_stratum eq 2.
� � � sample 135 from 615.
� � � else if unemployed_stratum eq 3.
� � � sample 145 from 900.
� � � else.
� � � Print /"oops! problem getting unemployment stratum", ID.
end if.
sort cases by unemployed_stratum.
compute ranorder = rv.uniform(0,2e31).
compute treat_group = mod($casenum,2).
value labels treat_group
� � � 0 'control'
� � � 1 'treatment'.

Art Kendall
Social Research Consultants


On 10/5/2011 9:37 AM, stace swayne wrote:

Hi Art,

�

Below is a response to your questions:

�

Have the subjects already been selected and you want to randomly assign them to treatment group being sure that each employment stratum is evenly divided?

The subjects have not been selected yet we are in the recruitment stage of the project.

�

Or are you going to draw a sample from each of three known populations? Or are you going to draw a large sample and post-stratify?

The goal is to recruit a large sample and then post-stratify.

How were the percentages in the strata chosen?� Proportional to stratum pop?� "over sampling" small strata to facilitate comparisons?

The percentages were chosen based on the interests of the community based organizations the primary investigator is working with. This is a small pilot community based study.

How did you decide on an N of 400?

The N of 400 was based on a power analysis as well as cost.

�

From: Art Kendall [hidden email]
To: [hidden email]
Sent: Tuesday, October 4, 2011 3:29 PM
Subject: Re: Some advice on SPSS sample selections

Please give more detail. for example:

Have the subjects already been selected and you want to randomly assign them to treatment group being sure that each employment stratum is evenly divided?

Or are you going to draw a sample from each of three known populations? Or are you going to draw a large sample and post-stratify?

How were the percentages in the strata chosen?� Proportional to stratum pop?� "over sampling" small strata to facilitate comparisons?

How did you decide on an N of 400?



Art Kendall
Social Research Consultants

On 10/4/2011 2:58 PM, John F Hall wrote:

Stace

�

I’m not a statistician and can’t really help you on this, so am copying this reply to the list where you will get better advice.� However, you need to give much more detail.� Also, have you thought about the SRMS section of the ASA?

�

�

John F Hall

�

[hidden email]

www.surveyresearch.weebly.com

�

�

�

�

From: stace swayne [[hidden email]]
Sent: 04 October 2011 15:08
To: John F Hall
Subject: Some advice on SPSS sample selections

�

Dear John,

�

� I need some guidance� on what type of syntax I need in order to do the following:

In my study there will be 400 subjects, 200 in a control group and 200 in a treatment group (people who will receive specialized job training).
�

My colleagues have recruited people who fall into three categories (1)� � recently unemployed six months or less, (2) unemployed for 12 months and� � (3) unemployed for greater than 12 months.

Members of my group are interested in having different proportions of the� � people who fall into the aforementioned categories represented in the treatment group. For example, 20 percent should come from the recently
� unemployed or 30 percent from the unemployed for greater than 12 months� � category etc.

Can� you give me some advice on how to do this via syntax?

Thank you,

�

Stace

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD