Colorectal Cancer Lymph Node Analysis
Budget: $30 – $250 USD
I have a retrospective colorectal-cancer dataset and want to understand which clinicopathological factors predict an adequate lymph-node harvest.
Your first task is to summarise the data: give me the mean, standard deviation and range for all continuous variables, plus frequencies and percentages for every categorical variable.
Once the descriptive picture is clear, move to inferential work. Build a binary logistic-regression model where the dependent variable is lymph-node harvest status (adequate vs inadequate). I need crude and adjusted odds ratios with 95 % confidence intervals, p-values, and a brief interpretation of each predictor. Please check model assumptions, report any multicollinearity diagnostics you run, and include goodness-of-fit output so I can judge the model’s reliability.
I am flexible on software—SPSS, R or SAS are all fine—just tell me which you plan to use and keep the syntax/scripts so I can reproduce the results.
Deliverables
• Cleaned analysis-ready dataset (if any recoding is required)
• Descriptive statistics tables
• Logistic-regression output with ORs, CIs, p-values and fit statistics
• Brief, plain-language summary of the key findings
• Script or syntax file and session log for full transparency
If something in the dataset suggests an alternative or additional test, feel free to flag it; otherwise, the scope above captures everything I need.
Descriptive Statistics /Mean, standard deviation (SD), range /Frequencies and percentages/Inferential Statistics /Binary Logistic Regression (The Predictive Phase)/Odds Ratio (OR)
Your first task is to summarise the data: give me the mean, standard deviation and range for all continuous variables, plus frequencies and percentages for every categorical variable.
Once the descriptive picture is clear, move to inferential work. Build a binary logistic-regression model where the dependent variable is lymph-node harvest status (adequate vs inadequate). I need crude and adjusted odds ratios with 95 % confidence intervals, p-values, and a brief interpretation of each predictor. Please check model assumptions, report any multicollinearity diagnostics you run, and include goodness-of-fit output so I can judge the model’s reliability.
I am flexible on software—SPSS, R or SAS are all fine—just tell me which you plan to use and keep the syntax/scripts so I can reproduce the results.
Deliverables
• Cleaned analysis-ready dataset (if any recoding is required)
• Descriptive statistics tables
• Logistic-regression output with ORs, CIs, p-values and fit statistics
• Brief, plain-language summary of the key findings
• Script or syntax file and session log for full transparency
If something in the dataset suggests an alternative or additional test, feel free to flag it; otherwise, the scope above captures everything I need.
Descriptive Statistics /Mean, standard deviation (SD), range /Frequencies and percentages/Inferential Statistics /Binary Logistic Regression (The Predictive Phase)/Odds Ratio (OR)