Analysis of Data for Psychology Dissertation (Please help)

Discuss statistics related things
Post Reply
grk99
Posts: 8
Joined: Tue Aug 09, 2022 5:03 pm

Analysis of Data for Psychology Dissertation (Please help)

Post by grk99 »

Hi,

I have attached an image of my data on Jamovi after gathering data from participants using an online questionnaire.

I am investigating the effect of pre-term birth AND low birth weight on psychosocial outcomes in adolescence/adulthood. Therefore, as you can see from the image, participants first disclosed if they were pre/full term, then disclosed the age they were born in weeks, then disclosed their weight at birth (I converted all to KG). They then completed a questionnaire on ADHD symptoms, life satisfaction, self-esteem and quality of relationships.

I am so confused what statistical tests to run with what variables, I seem to be going round in circles. I understand there will be several correlations involved, but can anyone shed any ideas on how they would deal with this data?

I have computed totals for each scale. But how do I deal with the first set of information regarding the pre-full term, weeks and weight?
Attachments
Screenshot (62).png
Screenshot (62).png (90.15 KiB) Viewed 8818 times
Last edited by grk99 on Thu Aug 18, 2022 8:00 pm, edited 1 time in total.
Bobafett
Posts: 76
Joined: Thu Jul 18, 2019 11:33 am

Re: Analysis of Data for Psychology Dissertation (Please hel

Post by Bobafett »

The lack of help from your University regarding the statistics seems a shame, but I guess as we are well into August you have passed the cut-off date for contacting your supervisor? There are a myriad of ways you could go about analysing your data: correlations, t-tests, linear / logistic regression, MANOVA etc.... I guess it all depends on what your hypotheses are? What is the research question you are trying to answer?
grk99
Posts: 8
Joined: Tue Aug 09, 2022 5:03 pm

Re: Analysis of Data for Psychology Dissertation (Please hel

Post by grk99 »

Hi,

Thanks for your reply.


Sorry - I should have included the questions I’m trying to answer!

• Is premature birth a risk factor for long-term psychosocial outcomes?

• Do gestational age at birth and birth weight status affect the severity of psychosocial outcomes in adulthood?
Last edited by grk99 on Thu Aug 18, 2022 8:11 pm, edited 1 time in total.
Bobafett
Posts: 76
Joined: Thu Jul 18, 2019 11:33 am

Re: Analysis of Data for Psychology Dissertation (Please hel

Post by Bobafett »

grk99 wrote: • Is premature birth a risk factor for long-term psychosocial outcomes?
You have a dichotomous grouping variable (full vs. pre) and so could compare these two groups on each single psychosocial measure you have used. If, as I suspect, you have multiple psychosocial DVs then you could perform a MANOVA instead and look to see if your groups can be separated across these measures.
grk99 wrote: • Do gestational age at birth and birth weight status affect the severity of psychosocial outcomes in adulthood?
These sound like correlations, possibly regression.

The details you mention make me think I know your supervisor. Please DM me.
simonmoon
Posts: 17
Joined: Thu Oct 27, 2022 3:45 pm

Re: Analysis of Data for Psychology Dissertation (Please help)

Post by simonmoon »

This article will help you and other community members on this issue.

MacCallum, Zhang, Preacher, & Rucker (2002). On the practice of dichotomization of quantitative variables. Psychological Methods, 7, 19-40.
User avatar
reason180
Posts: 274
Joined: Mon Jul 24, 2017 4:56 pm

Re: Analysis of Data for Psychology Dissertation (Please help)

Post by reason180 »

A good resource would be: https://www.learnstatswithjamovi.com/
ellen3539
Posts: 1
Joined: Mon Jul 03, 2023 5:22 am

Re: Analysis of Data for Psychology Dissertation (Please help)

Post by ellen3539 »

grk99 wrote: Thu Aug 11, 2022 2:19 pm Hi,

I have attached an image of my data on Jamovi after gathering data from participants using an online questionnaire.

I am investigating the effect of pre-term birth AND low birth weight on psychosocial outcomes in adolescence/adulthood. Therefore, as you can see from the image, participants first disclosed if they were pre/full term, then disclosed the age they were born in weeks, then disclosed their weight at birth (I converted all to KG).They then completed a questionnaire on ADHD symptoms, life satisfaction, self-esteem and quality of relationships.In this case, the involvement of a behavioral therapist could be beneficial. I found this behavioral therapist website on google, click this.

I am so confused what statistical tests to run with what variables, I seem to be going round in circles. I understand there will be several correlations involved, but can anyone shed any ideas on how they would deal with this data?

I have computed totals for each scale. But how do I deal with the first set of information regarding the pre-full term, weeks and weight?
Hi there!

I understand that you're feeling confused about which statistical tests to run with your variables and how to deal with the information regarding pre-term birth, weeks, and weight. I'd be happy to help you with that!

To start, it's great that you have already computed totals for each scale. This will simplify your analysis. Now, let's address the first set of information about pre-term/full-term status, gestational age in weeks, and birth weight.

For these variables, you can begin by examining their distributions and characteristics. It's important to understand the composition of your sample in terms of pre-term and full-term births. You can calculate the frequencies or proportions of participants in each group (pre-term and full-term).

To investigate the impact of pre-term birth and low birth weight on psychosocial outcomes, you can consider the following steps:

Group Comparison: Start by comparing the means or medians of the psychosocial outcome variables (ADHD symptoms, life satisfaction, self-esteem, and quality of relationships) between the pre-term and full-term groups. This will help you understand any initial differences between the two groups. You can use appropriate statistical tests such as t-tests or Mann-Whitney U tests, depending on the distribution and assumptions of your data.

Correlation Analysis: To explore the relationships between the first set of information (pre-term/full-term status, weeks, and weight) and the psychosocial outcome variables, you can perform correlation analysis. Calculate correlations between the variables using Pearson's correlation coefficient or Spearman's rank correlation coefficient if the variables are not normally distributed. This will give you an idea of the associations between these variables.

Regression Analysis: If you want to further investigate the effects of pre-term birth and low birth weight on psychosocial outcomes while controlling for other variables, you can conduct regression analysis. Consider using multiple regression analysis, where the psychosocial outcome variables are the dependent variables and pre-term/full-term status, weeks, and weight are the independent variables. This will allow you to assess the unique contributions of these factors in predicting psychosocial outcomes.

Remember, it's important to consult with your research advisor or a statistician who can guide you further and provide tailored advice based on your specific research questions and data. They can assist you in selecting appropriate statistical tests, interpreting the results, and ensuring the validity of your conclusions.

Don't hesitate to ask if you have any further questions or need additional clarification. Best of luck with your data analysis!
User avatar
patc3
Posts: 8
Joined: Thu Jun 15, 2023 3:25 am

Re: Analysis of Data for Psychology Dissertation (Please help)

Post by patc3 »

ellen3539 wrote: Mon Jul 03, 2023 10:18 am
grk99 wrote: Thu Aug 11, 2022 2:19 pm Hi,

I have attached an image of my data on Jamovi after gathering data from participants using an online questionnaire.

I am investigating the effect of pre-term birth AND low birth weight on psychosocial outcomes in adolescence/adulthood. Therefore, as you can see from the image, participants first disclosed if they were pre/full term, then disclosed the age they were born in weeks, then disclosed their weight at birth (I converted all to KG).They then completed a questionnaire on ADHD symptoms, life satisfaction, self-esteem and quality of relationships.In this case, the involvement of a behavioral therapist could be beneficial. I found this behavioral therapist website on google, click this.

I am so confused what statistical tests to run with what variables, I seem to be going round in circles. I understand there will be several correlations involved, but can anyone shed any ideas on how they would deal with this data?

I have computed totals for each scale. But how do I deal with the first set of information regarding the pre-full term, weeks and weight?
Hi there!

I understand that you're feeling confused about which statistical tests to run with your variables and how to deal with the information regarding pre-term birth, weeks, and weight. I'd be happy to help you with that!

To start, it's great that you have already computed totals for each scale. This will simplify your analysis. Now, let's address the first set of information about pre-term/full-term status, gestational age in weeks, and birth weight.

For these variables, you can begin by examining their distributions and characteristics. It's important to understand the composition of your sample in terms of pre-term and full-term births. You can calculate the frequencies or proportions of participants in each group (pre-term and full-term).

To investigate the impact of pre-term birth and low birth weight on psychosocial outcomes, you can consider the following steps:

Group Comparison: Start by comparing the means or medians of the psychosocial outcome variables (ADHD symptoms, life satisfaction, self-esteem, and quality of relationships) between the pre-term and full-term groups. This will help you understand any initial differences between the two groups. You can use appropriate statistical tests such as t-tests or Mann-Whitney U tests, depending on the distribution and assumptions of your data.

Correlation Analysis: To explore the relationships between the first set of information (pre-term/full-term status, weeks, and weight) and the psychosocial outcome variables, you can perform correlation analysis. Calculate correlations between the variables using Pearson's correlation coefficient or Spearman's rank correlation coefficient if the variables are not normally distributed. This will give you an idea of the associations between these variables.

Regression Analysis: If you want to further investigate the effects of pre-term birth and low birth weight on psychosocial outcomes while controlling for other variables, you can conduct regression analysis. Consider using multiple regression analysis, where the psychosocial outcome variables are the dependent variables and pre-term/full-term status, weeks, and weight are the independent variables. This will allow you to assess the unique contributions of these factors in predicting psychosocial outcomes.

Remember, it's important to consult with your research advisor or a statistician who can guide you further and provide tailored advice based on your specific research questions and data. They can assist you in selecting appropriate statistical tests, interpreting the results, and ensuring the validity of your conclusions.

Don't hesitate to ask if you have any further questions or need additional clarification. Best of luck with your data analysis!

You literally copy-pasted this from ChatGPT lol... The answer seems OK, but you should definitely at least cite it
User avatar
reason180
Posts: 274
Joined: Mon Jul 24, 2017 4:56 pm

Re: Analysis of Data for Psychology Dissertation (Please help)

Post by reason180 »

RE: "You literally copy-pasted this from ChatGPT lol... The answer seems OK, but you should definitely at least cite it"

Perhaps we have BOTs participating in this forum?
Post Reply