I will do python data analysis, statistics and clear data visualization
Python Data Analyst, Rigorous Statistics, Qiskit and Quantum Computing
About this Gig
I analyse survey data and datasets in Python, and I tell you which results are real.
WHAT YOU GET
- Your data cleaned and checked: missing values, duplicates and impossible entries reported, not silently dropped.
- The right test for your question, with the assumptions checked and stated.
- A clear chart for every key result (Standard and In-Depth).
- A plain-English summary you can hand to a supervisor or a client.
WHAT MAKES THIS DIFFERENT
Most reports present everything as a finding. I compare your results against what you would see by chance anyway, so you learn which effects are solid and which are noise. If a result does not hold up, I will tell you before you present it to someone who checks.
I work with survey exports (Google Forms, SurveyMonkey, Excel, CSV), experimental data and any tabular dataset. Python, pandas, NumPy, SciPy and Matplotlib.
BACKGROUND
I built and ran a 2,031-experiment benchmarking campaign on IBM quantum hardware, with a verification script that recomputes every reported number from the raw data. It is in my portfolio below.
Send me your data and your question before you order, and I will tell you honestly whether I can answer it.
My Portfolio
FAQ
What file formats are accepted?
CSV, Excel, Google Sheets, SPSS .sav, and exports from Google Forms, SurveyMonkey, Qualtrics and Typeform. If you have something else, send it and I will tell you before you order.
My data is messy and has missing answers. Is that a problem?
No, that is normal. Cleaning is included. I report what I removed and why, rather than quietly dropping rows, and missing data is handled explicitly and stated in the report.
Can you deliver SPSS or Stata output?
No. I work in Python, and I would rather say so than fake it. You get charts, tables and a written report. If your supervisor specifically requires SPSS output, I am not the right seller for you.
What if my results are not statistically significant?
Then I tell you clearly, and explain what sample size would be needed to detect an effect that size. A null result is a real result. Not dressing one up as a finding is the point of hiring me.
How many respondents do I need?
It depends on the size of the effect you are looking for, not on a magic number. Message me your question before ordering and I will tell you whether your sample can answer it, or what it can answer instead.
Do you write survey questions or code open-text answers
Not in this gig. I analyse quantitative responses. Question design and thematic coding of free text are different skills, and I would rather point you elsewhere than do them badly.
Can you help with my thesis or dissertation?
Yes, with the analysis and with understanding your own results well enough to defend them. I do not write dissertations or produce work to be submitted as your own unaltered.
Will I get the code?
Yes, on request, as a commented Python notebook so you or anyone else can rerun the analysis and get the same numbers. Reproducibility is included, not an upsell.
Is my data kept confidential?
Yes. Your data is used only for your order, never shared or reused, and deleted once the order is complete if you ask me to.

