Statistical Data Analysis - Science topic

How can design of experiments optimize reaction conditions in catalytic processes using statistical data analysis?

📷📷Elevate Your Research with DataStatPro📷📷 https://datastatpro.com

Discover precision and simplicity in statistical analysis. DataStatPro empowers researchers, academics, and professionals to transform raw data into actionable insights through an intuitive, browser-based platform.

Why DataStatPro?

📷 Comprehensive Toolkit

From basic descriptive stats to advanced regression models, access 50+ statistical methods tailored for academic rigor and real-world applications.

📷 Universal Access

Skip downloads—launch advanced analyses directly from any browser. Compatible with Windows, macOS, iOS, Android, and Linux systems.

📷 Guided Workflow

Our step-by-step interface automates assumption checks (normality, homoscedasticity, multicollinearity) while educating users on methodological best practices.

📷 Publication-Ready Outputs

Generate clear English interpretations of results, complete with APA-formatted statistical reporting (β coefficients, p-values, effect sizes).

📷 AI Analysis Assistant

Get real-time guidance on analysis selection, data cleaning, and result interpretation.

Launch Your Next Project

Visit https://datastatpro.com

#StatisticalAnalysis #ResearchInnovation #DataScience

Why is it important to carry out statistical data analysis of signal - to - Noise Ratio in an MRI and CT scans?

I am conducting an umbrella review, a review of systematic reviews and meta-analyses, where we pool relevant outcomes from distinct meta-analyses.

The problem is that I do not know which software can do this type of analysis!

I appreciate your help with this issue.

Regards, Ammar.

For example in the field of Engineering, Medicine, Psychology, Business etc...

I have done different treatments on different samples for some time, and I have collected data from different analyses, if I want to perform ANOVA, which ANOVA to choose and how to index it in graph

Some of my students are dealing with Financial Literacy, Quality of Life, DASS-21 and CES-D. I have seen that most papers present descriptive, basic comparison, correlation, GEE and SEM.

Focusing on implementation of survey response scale (4/5/7/10 Likert), what tool would you recommend for the analysis ? (statistical data analysis, machine learning, optimisation, metaheuristic, etc.)

Thank you for your suggestions 🙏

I would like to hire a statistician to analyze epidemiological data: calculate odds ratios, chi-squares, perform regressions and advise on whether there are other analyses that can be done with the data. Can you recommend someone?

We usually use Excel in our lab to analyse data, but I would like to take a course on a more sophisticated tool. Can you share which one is most common in molecular biology?

Thank you :)

I came across a commentary titled, 'Tests of Statistical Significance – their (ab)use in the social sciences' and it made me reflect on the validity of using my sample for statistical testing. I have a sample of 24 banks and they were not randomly selected. They were the top 50 banks ranked by the Banker and I narrowed down the sample to 24 because only those banks were usable for my study. I wanted to test the association between these banks using a McNemar's test but any result I obtain- I obtained insignificant results - would be meaningless, right? Because they are not a random selection. I did not want to make a generalisation, but I wanted to know if I could still comment on the insignificance of their association?

I have some data-sets causing me and my limited statistical knowledge problems.

When handling qPCR values or quantifications of WB, you always "normalise" your values to a endogenous control / loading control i.e. a house-keeping gene. And then afterwards often to a control treatment as well.

However, this means the data is now fold change compared to a control. So all values less than control is between 0 and 1, and all values higher than the control is 1 to infinite. This heavily skews the data. To my understanding a parametric test like a simple t-test or ANOVA cannot be used, as it will it would handle fold change values 0.1 (A) and 1.9 (B) as absolute values, being equally apart from 1 (C), even though A is actually ten times less than C, while B is only 1.9 more than C.

I have a dataset from three replicate experiments, where I have transfected with either control siRNA or 3 other siRNAs (1-3) (4 samples). 

I have subjected each replicate experiment to WB one at a time. In the WB I have blotted for 3 different proteins and a loading control (4 proteins). Afterwards I have quantified the WBs. 

I.e 3 biological replicates, with 4 treatments, having 4 readouts (normalised to the same loading control).

In Prism, I have done two-way ANOVA on the values normalised to loading control, where I compared each treatment to the control treatment, using Bonferroni correction. 

I have tried to do the test on both log transformed values and not log transformed data. I would guess it would be most appropriate to do the test on the transformed data.

However, the treatment with siRNA 3 results in small increases of some of the proteins with little variation between experiments, while another siRNA 1 results in decreased values but with big variation. When I do the ANOVA test on non log-transformed values the treatment with siRNA 3 is significant for these proteins, which seems right looking at the values and CI for both log-transformed and not transformed data. But when I do the ANOVA test on the log-transformed data it is no longer significant, while treatment with siRNA 1 (which had high variation), suddenly becomes much more significant. This does not seem right to me. 

So my questions is should you always log-transform qPCR and WB data (fold change data) before doing a statiscal test? Or is their some problems in transforming your data, that will result in e.g. false negatives?

somon four group design is an experimental design contains

(a) a treatment group that receives both a pretest and a posttest

(b) a treatment group that receives a posttest only

(c) a control group that receives both a pretest and a posttest

(d) a control group that receives a posttest only.

Given a time series with Events, I want to test whether events in two time series are occurring differently. 

See for example the attached image. There are 12 events (orange) between 2000 and 2007 with different lengths. Let's pretend this are drought periods in countries or a any kind of event that can last for a certain time and there can be multiple events overlapping (my actual dataset are returns and the events are some kind of patterns).

I can simulate data to generate events. For example I can simulate a weather dataset and if certain conditions are met it is an event that will last for some time (e.g. no clouds = no rain). Thus, I can generate as many datasets (under H0) like I want. 

I want to check if there is any kind of "systematic occurrence" or if there is the same amount and distribution of events in the simulated datasets and actual dataset. 

I could use a simple t-Test and test the average number of events from the simulated datasets against the actual one. But his would not account for the problem, that the amount of events could be the same - but the events could be clustered (e.g. the twelve events in the actual dataset are always at the beginning).

Does anyone knows any similar problem with a solution or any kind of test for this kind of problem?

My only idea is to split the tests:

(1) Test for the number of occurrences 

(2) Somehow test if the structure is different

Thanks,

Nico

Dear scholars,

I have a database with 100 geolocated samples in a given area, each sample contains 38 chemical elements that were quantified.

Some of these samples contain values Below the Detection Level of the instrument (BDL), clearly, when we have 100% of the samples with BDL values there is not much to do, but what can be done when, for example, when there is only 20% BDL, what do we do with them, with what value do we replace a BDL sample?

Some papers show that a BDL sample can be replaced by the detection level (for the instrument's minimum detection level for that chemical element) divided by 0.25, others show that you have to divide it by 0.5... What would you do in each case, and is there any literature you would recommend? If it matters, I am mostly interested in Copper and Arsenic.

Regards

Dear scholars,

I have a database with 100 geolocated samples in a given area, each sample contains 38 chemical elements that were quantified.

Some of these samples contain values Below the Detection Level of the instrument (BDL), clearly, when we have 100% of the samples with BDL values there is not much to do, but what can be done when, for example, when there is only 20% BDL, what do we do with them, with what value do we replace a BDL sample?

Some papers show that a BDL sample can be replaced by the detection level (for the instrument's minimum detection level for that chemical element) divided by 0.25, others show that you have to divide it by 0.5... What would you do in each case, and is there any literature you would recommend? If it matters, I am mostly interested in Copper and Arsenic.

Regards

For Individual responses I can calculate the value with respect to which we have to check the outlier don't know?

Please help in this regards

I am looking towards learning the statistical analysis involved with the linkage of air pollution with health endpoints. The fundamental knowledge about time series analysis and epidemiology is critical. Would appreciate if resources can be shared about statistical data analysis

Hello SEM experts,

I have a 2nd order reflective construct which have 6 first order factors all having 2 questions each. Can i combine those two questions for each 1st order factor by using item parceling approach for CFA and SEM? Actually i am trying to reduce the number of observed variables because of so many questions in my questionnaire. What are the implications if i do this?

I am using IBM AMOS for CB-SEM.

Hi, I am an undergraduate currently working on a project that is using a quantitative survey.

I have developed 3 scenarios that have the same 5 Likert scale questions across these scenarios. Also, the questions are split into confidence and experience, as they are asking respondents to self-rate themselves on confidence and experience on the skills specified in the questions.

My question is, how should I analyse the Likert scale responses across all 3 scenarios? Can I sum them up then divide them to get the mean value of each response to each question? I cant seem to find similar papers like my situation.

I have found Cronbach alpha to be >0.7 across all the questions and there is significant positive correlation between confidence and experience across all 3 scenarios. Are these valid reasons enough to add up arithmetically the responses across the 3 scenarios? I can't find any research to say when I am able to add the responses up.

Please help as I am quite lost. Please cite sources in your statement so I can read up further too.

In vitro experiments ，there are only 6 samples per group, what method is used to compare the differences between groups? I want to use proper statistical data analysis method.

Hi,

I have RT-qPCR data that I am analysing currently, but am unsure of how to correctly express the statistical analysis of it.

The experiments (two independent replicates) contain the following;

Control sample, test sample

Endogenous control gene, 6x target genes

I want to express the data as fold change, but need to add error bars and statistically analyse for significance. I have thus far calculated the delta delta CT value for each set of samples for each target gene relative to the endogenous control, and have calculated fold change (=2^-ddCT) of the test samples relative to the control samples.

Do I use the Std Dev of the ddCT values as the error bars, and how do I analyse for significance? (my thought was using test versus control samples with independent samples T test, but with what values?)

Thanks in advance

The study consists of three variables - IV, mediator and DV. Data was collected three times with a lag-time of one year between each wave (longitudinal survey data). I intend to kind of establish causality? How to do analyze it and can I use AMOS or Hayes Process for this purpose?

Hello,

In relation to another recent, as yet unanswered, post (https://www.researchgate.net/post/How_to_calculate_comparable_effect_sizes_from_both_t-test_and_ANOVA_statistics), I am wondering how I can calculate the sample variance of an effect size, in order for me to then infer the confidence intervals.

So far, I have been calculating 'Cohen's d' effect sizes from studies' experiments, by using the t-value of two-sample t-tests and the sample size (n) for each group. I then convert the Cohen's d into an unbiased 'Hedge's g' effect size.

I understand that, normally, in order to calculate sample variance, I would also need to know the means of the two groups and their standard deviation. However, these are not reported in most studies when a t-test is calculated. Is there any way I can calculate the sample variance of my Hedge's g effect sizes without this information?

Many thanks,

Alex

The mentioned test is not conclusive because the results of this test validates the exogenous instruments even most of them are endogenous. So, I will be greatly appreciating any suggestions.    

In the simulation I am conducting, I have a set of triangles and I select the optimal triangle based on my metric. After every simulation, I obtain an optimal triangle and I note down the lengths of its three sides. So i conducted this simulation a large number of times and noted the lengths of the three sides of the optimal triangle. Now I want to see if there is any pattern which these three sides follow or if there is any relationship among these three sides, or even between two sides.

For example, the names of the three sides are D1, D2, and DX, a relationship can be something like , D1=D2=Dx or D2=2 times Dx, etc. (simple examples, it may have a little complex relation).

I have seen that D2 and DX have a close correlation, how can I use this to relate it with D1 OR what conclusion can I draw from it? How can I analyze these three variables together? Whether I can do something like D1 Vs (D2 and DX)?

The dataset looks like this: (but having large number of data)

D1 D2 DX

3.3 3.0 3

1.12 0.83 1

4.4 6.44 6

2.5 11.9 15

0.79 2.19 3

3.81 9.63 12

2.4 4.9 7

2.56 13.2 13

1.97 5.26 5

4.50 3.68 4

4.72 5.21 6

0.89 4.10 5

1.78 5.89 8

3.2 7.54 12

3.18 2.20 4

My study model has two Independent variables, one dependent and one mediating .

How to assess collinearity in PLS-SEM?

I have a very simple model where I measure the effect of tone of voice on purchase intention, moderated by brand alignment. I have a 2x2 between subjects model, with tone of voice levels being informal vs formal, and brand alignment levels being warm vs competent.

When I run a two-way anova, I see that the main effect for tone of voice is not significant. However when I run it in a one way ANOVA, this effect is significant. Can someone explain why this is, and if i were to report the one way anova would this be incorrect?

for bachelors thesis

I am straggling to use SPSS amous software. Can you give some suggestions to me?

Apart from using Microsoft Excel to draw graphs, which other software is user-friendly that I can use?

How does oversampling technique like ADASYN, MWMOTE handle the nominal, or categorical features in dataset

We are conducting a study about the correlation of covid-19 psuedoscience theories to the proliferation of COVID-19 positive cases. Which statistical test should we use to correlate the data that we are going to gather from the Covid-19 pseudoscience theories scale (4pt likert scale) to the number of positive cases? (data of the covid-19 positive cases will be from the health departments tally). Thank you for those who could help us in advanced. ☺️

Dear Scientists

During a study of corrosion inhibitor's performance, and doing Tafel fitting, there is a statistical value appear called "Chi Squared" and having a wide range of values ranging from very low (0.0001) to tens. 

Please can you advise how can I decide based on "Chi Squared" value if my Tafel fit is correct or not? 

I am using GAMRY Potentiostat. 

Thank you in advance

Aasem

Hello everyone, I would like to ask if the way that the sample size of this research was calculated is valid or correct. Is a study to evaluate the effect of gargling with Povidone iodine among COVID-19 patients. The text says “For this pilot study, we looked at Eggers et al. (2015), using Betadine gargle on MERS-CoV which showed a significant reduction of viral titer by a factor of 4.3 log10 TCID50/mL, and we calculated a sample size of 5 per arm or 20 samples in total”. From this data of the reduction of the viral titer in a previous study on MERS-CoV ¿It is valid to calculate the sample size this way for a new study on COVID-19?

These tools are powerful and allow for a wider and deeper understanding of datasets, but which is the best one for Social Economics? The criteria for this comparison is set by:

a) How easy is it to learn?

b) How easy is it to use?

c) Tools (repositories, etc.)

d) Cost

Hi everyone

I have recently conducted an experiment

my experiment design is like below:

group 1 T1 X T2 T3

group 2 T1 X T2 T3

group 3 T1 X T2 T3

group 4 T1 X T2 T3

now whats my best option to analyze the collected data in SPSS?

I have used the repeated measures in general linear model section, but it has its shortcomings and don't give me the data I need.

Dear to whom it may concern,

I would like to ask you about the normalization methods used to remove non-biological variations from the metabolomics data.

Because there have been many normalization methods reported until now, I am so confused about based on what criteria to select the best normalization method for a particular metabolomics data? Also, what is the meaning of each normalization method?

I hope that you may spend your little time clearing my questions and if convenient for you, may you show me the documents or tips in this case, please?

Thank you so much,

Pham Quynh Khoa.

I measured call rates for manatee vocalizations and have a range of call rates from 0-11 calls/min. The data set for all observations are not normally distributed, however I want to report call rates for individual groups (ie. Call rates for one manatee, groups of two, etc). Is it more appropriate to use the median and interquartile range to describe each set or would the mean and standard deviation be acceptable. I do not have a large standard deviation when looking at individual groups. Ie. (Call rates for groups of 6 animals was 6.28 calls/min with a standard deviation of 4.21)

Hello, I have a query

What happens is that I have some data (continuous), that theoretically should have some kind of relationship, and I want to make some statistical test of association between my data (which are few, les than 10 samples for some groups) and other data set (Other features of interes)t which are many. The problem is that, I try doing random data based on the means and standard deviations ( on mintab and using the normal distribution) but shows no  association or dependency  type , probe with regression and correlation (nonparametric too), what kind of approach should I use in these case?

thanks for the help

I have analysed my data and have a significant H- value. But I would want to see where my samples are different. Could someone please help me on how I can be able to do post-hoc test after Kruskal-Wallis in STATISTICA. I would like to do a Tukey HSD.

Dear seniors and researchers,

I am currently performing ROC curve analysis in SPSS v22 to asses the diagnostic value of several laboratory tests to predict parasite density status (whether hyperparasitemia or not) of patients with malaria. I found that investigated parameters I examined showed a high AUC while the p value is not significant. Regarding the circumstance I faced, what are possible factors that might influence the significance of AUC in ROC Curve analysis? Does the number of samples contribute to this finding?

Thank you

When calculating required number of subjects in designing research protocol. Three parameters, significant level, power and effect size, are required. Significant level (P<0.05) and power (=0.8) are easy to set. For the effective size, I am referring Cohen’s review “A power primer” (PMID: 19565683). The review shows the effect size indexes for SMALL, MEDIUM and LARGE in various tests in Table 1. But the table does not show the values for paired t test. Please let me know the values and the publications I can refer to.

The research question is: Does gender influence stress levels four month post-diagnosis of cancer?

I am making a research design for estimating the relation of three independent variables namely organic, conventional and integrated agriculture in relevance to the decrease of soil erosion. i need to decide on what kind of analysis i need, that is, bi-variate or multivariate data analysis.  i feel that taking pairs of correlation between pairs of variables is more appropriate than looking for interaction that leads to a aggregate measurement that is, multiple correlation or multiple linear regression. so anyone that can comment and help to answer my question will be appropriate as a feed back.   

Hi there,

I do not have a solid understanding of the statistics - I am seeking advice regarding the correct data analysis. I am helping my colleague to analyze the oxidative stability of oil. Please help us!

This is what our data looks like

We checked oxidation products in oil samples in triplicate (n=3) at 6 different time points (0, 2, 4, 6, 8, and 10 weeks) under 2 different relative humidities (11 vs 44%).

I am confused as to which statistical test(s) would be the suitable one.

I really look forward to hearing from you awesome scientists:-)

I have predicted crop yield as a function of two treatments, each treatment at four levels. The model does not work well, so I use a log transformation. I then calculate least square means and run a Tukey test on these means. I plot a nice graph ...... and get complaints. A common suggestion is to back-transform the results. This seems simple enough but I don't know what the numbers mean. The problem started with typing in a bunch of numbers into Excel. I can then take the mean of these values. I can take the log of each value. I can back-transform the mean(log(value)) and find that it is nothing like the mean of the untransformed values. The cause is that the log transformation changes the distribution of the data. Needless to say back-transforming the LSMeans and SE in the original problem did not seem to work very well either.

   There are solutions. For example, present the raw means and standard deviations, along with the multiple comparison results from the model with the transformed data. I have no problem with that.

   I can find statistics textbooks that recommend back-transformation. So I was wondering if it was generally accepted. If so, then how does one interpret the result? It seems to me that applying the term "mean" would be misleading, and I do not see figures with labels like "back-transformed(log(value))."

Hello,

My research will have 2 groups (intervention & control groups) comprised of couples in a romantic relationship. I will sample each participant three times for a period of 45 days (at the beginning of the study, after two weeks and a month after that), so time as a within subject factor.

What do you think is the best analysis for this study design?

Any help will be appreciated.

Thank you in advance!

There are several ways to validate quantitative research findings with the help of different programs like SPSS and other stat tools. I want to ask as if there is any method to validate or proof the results of qualitative research findings like interviews and how to include this qualitative date in the research article.

Hello,

I am currently undertaking a meta-analysis related to the efficacy of previous studies to harness pharmacological agents to ameliorate associative fear memories in rodents. The concept of effect sizes is fundamental to meta-analyses, but I am not so familiar with it. As a result, I have a (perhaps) elementary question related to the calculation of effect sizes:

I have decided that 'Hedge's g' will be the most appropriate metric of effect size for my analysis, given the relatively small sample sizes of my included studies. Often, authors will report an observation-of-interest (i.e. treatment vs. control effect) in the form of a (two-sample) t-test. I have found that the reported values of a t-test can easily be used to calculate a Hedge's g effect size - all good so far.

However, often an observation-of-interest is reported as an ANOVA (e.g. when the treatment group is compared with more than one control type), and other meta-analyses I have read seem to derive their effect sizes from both t-test or ANOVA statistics, depending on which is carried out. Again, I think I have found the correct equation to derive a Hedge's g effect size from the reported F-value of an ANOVA. However, I am struggling to understand how effect sizes can be comparable when they are derived from both t-test and ANOVA statistics. Surely ANOVA-derived effect sizes are less informative because you cannot tease out the individual contrasts-of-interest from an F value, as you obviously can for a two-sample t-test.

As a result, my initial instinct would be to request the raw data from the author (when ANOVAs have been reported), calculate a t-test for the specific contrast-of-interest (i.e. treatment vs. one particular control group) and calculate the Hedge's g effect size from the resulting t-test value.

Apologies for the long question, but I would hugely appreciate someone to enlighten me, and show me how to conflate t-tests and ANOVAs when generating effect sizes for a meta-analysis.

Many thanks!

Alex Nagle

I have data from an assessment tool administered to two samples from two sifferent countries. Can somebody help me with the best statistical method to assess the cross-regional validity of the assessment tool?

Does the modern testing theory help solve the misleading nature of drawing conclusions based on the significance or the non-significance of the P-value?

Do you use tools like SIMCA or MetaboAnalyst in your work?

I'm part of a team at Kitware (https://www.kitware.com) attempting to address some major issues metabolomics researchers face with a new free and open source platform for statistical analysis of metabolomics data.

This tool is being funded by the US National Institute of Health; Grant details can be found at: https://projectreporter.nih.gov/project_info_description.cfm?aid=9792042&icde=43584022&ddparam=&ddvalue=&ddsub=&cr=4&csb=default&cs=ASC&MMOpt=

In particular, we are looking for:

Please feel free to either reply through these forums or email brandon.davis@kitware.com. If you are available to follow up and tell us more about your lab's needs via video chat, please email me and I can schedule something with our team.

Thank you!

Dear Experts,

I am in the process of writing my master thesis on the factors that influence the intention to purchase a sustainable children's product. I do not have any knowledge in statistics and my mentor does not seem to know as well. Thus, I am resorting to online resources in the hope to find help. My conceptual framework consists of 5 constructs that may or may not influence the intention to purchase green; namely, attitudes, subjective norm, perceived behavior control, environmental knowledge, product attributes.

In the survey, I measure each construct with 2-4 Linkert Scale questions with 5 possible answers from "Strongly Agree to Strongly Disagree" (total of 18 questions). Additionally, I added demographic questions.

I would appreciate any feedback and I wish everyone stays safe and healthy.

If I need to find out a suitable sample size population to utilise in a chronic disease case control study, how do I use an allelic frequency derived from a previous study on a chronic disease prevalence to calculate this value.

T-test is usually applied on interval data. Is it okay to apply on Likert data? How?

I would like to conduct Monte Carlo uncertainty analysis for actual and predicted data in R;

Kindly share some effective resources;

However, I have this <<https://rdrr.io/rforge/metRology/man/uncertMC.html>> where understanding the hyperparameters of it is another challenge.

Please share your views/knowledge/ effective stuff/materials

best regard!

One of my student has performed a statistical analysis based on data obtained from Likert scale. However interpretation from t-test  which is based on the null hypothesis that the means are same and p <.05 and hence hypothesis that the statements are complied with ...(means are same) has been proved?!!! Hope this is wrong.

I have a set of nonlinear ordinary differential equations with some unknown parameters that I would like to determine from experimental data. Does anybody know of any good freely available software, or good reference books?

I have done a questionnaire about the impact of social media on sales and I have got 91 responses out of 120 and I want to analyze it by using statistical data analysis to support my hypothesis.

I am working on "Causes of students' failure".

I distributed a questionnaire to students and some variables were tested to find out if they are responsible for students' failure in English classroom.

Some of those variables are lack of motivation, anxiety, inappropriate study methods, teachers etc.

Are these variables dependent or independent?

Thank you very much indeed!

I am doing my dissertation on electrical smart meters design. I am questioning whether the design of the features is limiting the potential Reduced Household Energy Consumption (RHHEC) as currently, smart meters are underperforming in terms of RHHEC when compared to predictions.

I have data on energy reduction status binary Yes/No and a 5-point Likert scale from a large decrease to a large increase in energy consumption. I also have data on 12 different smart meter features concerning:

I have supplemented these questions with comment boxes to collect information on what makes an easy to find and use a feature and what households valued in a useful feature

What I want to do is compare this information between those households that saw RHHEC and those who did not to look for trends e.g. households who find the features hard to use or cannot find a certain do not see RHHEC

How would I go about analysing these?

I have considered a chi-square, linear regression or 2 samples t-test, can I perform multivariate analysis on all 12 features at once compared to RHHEC status (Y/N) or should I assess each feature individually?

Following this can I perform PCA or use dendrograms or is a simple scatterplot matrix the best?

Hello researchers!

I am currently writing my thesis and is trying to analyse my data. We thought we could use a z-test but since our collected data is not normally distributed we are trying to find a new method to use. We collected data via a survey where the participants looked at prototype 1 and later answered questions in a 5-point Likert scale, before moving on to prototype 2 and answered the same questions for that one as well. We want to compare the two prototypes as well as look at them individually. The participants were a convenience sample and we got about 80 answers.

Does anyone have a suggestion on what kind of data analysis method we could use?

Any help is very welcome, thank you!

Hi all,

I'm looking at accelerometer data. I used two accelerometers at once for a backup, but now I need to correlate the two datasets provided. I'm only looking at one axis on each, but one set of data has ~5000 more data points than the other. This may sound like a lot, but the datasets on average hold about 215,000 points of data. During data collection, I paired the two so they began collecting data and ended at the same time, but still one observed more movements than the other. Is there anyway around this to compare the two to say that both pick up about the same type of data than the other? A correlation test in R isn't working since there are two different amounts of data in each column. I do, however, have the mean and SD of each set.

Thanks so much in advanced.

I have two datasets containing Z-score data. However both of them are of highly different sizes, i.e one has 500 samples while the other has 4. I want to determine if the differences between the samples is statistically significant. I thought of using an unpaired t-test, but I am not sure if the difference in sample sizes would invalidate the method.

I need to compare two groups with equal /unequal variances. Which t-test is the best? If I am using Welch's unpaired t-test, should I do an 'F-test two-sample for variances' first and see if the groups have an unequal variance?

Thank you for your answers

Most of the young researchers get confused with these two statistical tests while analyzing. In order to better understand this issue, I am putting forward a discussion for expert opinions. When is the chi-squared test not the exact fishing test to be carried out? and when to perform fisher's exact test instead of the chi-squared test?

I am a bit stuck with the following problem. I am currently performing a meta analysis on observational studies using CMA where my outcome variable is the performance on a cognitive test as a function of adherence to an exposure variable (on a scale of 0 to 9). Normally,  the results are presented either as means differences between tertiles according to the level of adherence (low, middle or high) or as a regression coefficient per additional unit of the exposure variable.

My main question is, how can I pool together both types of studies into the same meta analysis? I have found a similar question on risk estimates that suggest to estimate the linear trend of the categorical results. I don't have access to raw data and I only know sample sizes, mean differences and confidence interval. Is it possible to do the same in this case? If so, how should I do it?

I was thinking of just including each tertile comparison as a subgroup of the same study, and leave the continuous variables as they are. But I am not sure if this lousy approach is acceptable. 

Thanks.

Hello,

I need some advice about how to statistically analyze WB data.

I am trying to compare the levels of two proteins in mutants relative to wt samples (it's mouse embryonic tissue).

I repeated the blot 3 times, with control and mutant samples in each run following exactly the same protocol each time. (Because of the type of sample, it's not easy to increase the number of blots or to run all samples in the same blot, which would be better for analysis). Then, for each blot I obtained the relative intensities and normalize with Actin to get the normalized relative intensities (NRI) for the wild type and mutant samples.

Visually (and checking the NRI), the differences are very obvious but I would like to see if the differences are statistically significant.

So my main question is about the statistical analysis of these data. After looking for information and not finding much, what I have tried is to use Linear Mixed-Effects Models with genotype and protein (plus their interaction) as a fix effect, and run as a random effect, then do an ANOVA and calculate the pair-wise differences between groups. Run is not significant and the level of one of the protein is, which is consistent with what I see in the blots.

Is this analysis correct? Does anyone know any other proper way (if there is any) to analyze these data?

Thank you in advance

I have individual level data from several rounds of surveys. These individual responses are nested within sites. I have other data measured just at the site level. However, there are only 10 sites, and I have a number of factors (more than 5) I would like to turn into a composite score. Is there a way to create a composite score for sites with so few cases (N=10) and use that to predict average individual responses?

Hi, I'm running an antibody-dependent phagocytoisis assay, where we add serum to fluorescent antigen-coated beads, then donor neutrophils, and subsequently measure the amount of phagocytoisis as MFI (fluorescence) within our neutrophil population to give us a phagocytosis score.

As we run the assay on different days with different neutrophil donors we are also running a titration of a serum standard. I was wondering how best to utilise this standard curve?

One method we're using is similar to what you would do with an ELISA, whereby you interpolate the values of our diluted sample from the standard curve. But this is only giving me a somewhat arbitrary value of the dilution required (which I can convert to phagocytic units, but again this is just done arbitrarily). The other method I've tried is min max normalisation using the top and bottom of the standard curve values.

What I would ideally like is a method of normalisation that gives an output as phagocytic score, as this seems biologically most relevant. Any input is appreciated.

Cheers,

Mari

Hi

Historical data is used to forecast the number of functional failures in passenger trains. However, there is always differenc in forecast and actually observed data. Am wondering which technique or approach is suitable to minimize the forecast error in case of railway data.

Shall be greatful if you can share some useful article or case study.

Hello dears,

In regression modelling process, somtimes we deal to make a categorization of a continuos variable ( DVs or IDVs), What are really the potential problems inherent of such transformation, on:

Thank you so much for any response and clarification

Hi, I have conducted a literature review on performance of warm mix asphalt containing RAP mixtures. I tabulated several properties such as Rutting, low temperature cracking, fatigue etc. and number of studies showing positive results and negative for each property. For example, if 20 studies were selected for fatigue, there are 15 studies that shows positive and 10 that shows negative. I want to come to a conclusion based on this result whether WMA containing RAP is performing better in Fatigue or not. How can I statistically prove this? Which statistical tool should I use? Thanks in advance.

I'm looking for a free and user-friendly tool. I'm familiar with python (and a bit R).

Thank you

I have a little question and need some clarification on chi square's strength of association (using the Phi & Cramer's V). Say I have 4 parameters, all of them show significant relationship with the e.g., Quality i.e., p<0.05 in chi square test, but I want to prove more results by knowing the strength of each parameter, hence I use the phi and cramer's V, then I compare all the four parameters' association strength. From this, I know which one of them has the strongest relationship (most influential) as well as the least. Therefore, I would like to know whether or not such comparison between the parameters' strength is valid and allowable?

Thank you and your comment is highly appreciated.

HI RG colleagues,

some statistics' comments believe that the Baron And Kenny 1986 mediation method out of date. they suggested new effective approaches in term of testing the mediating variable.

Hi Colleagues,

For example. I have a data set which can be categorized based on sampling time and treatment. However, I only concern the main effect of treatment on parameters I measured. What I did is to conduct one-way ANOVA with treatment as the only factor to analyze data for each sampling time, respectively (e.g., day by day).

I want to know that is there any rules ask us to must conduct or report the two-way ANOVA before a one-way ANOVA even we don't want to discuss in the paper? I only want to enhance the readability and stress the effect of treatment.

Thanks.

Best,

Guijun

Do anyone know how to perform a linear regression with averaged data instead of individual data points? If I have, for example, data on a certain parameter for a number of species (let`s say photosynthetic rates) and, after a while, I go and collect data on nutrient concentration for the same species - but obviously different individuals - is it possible to correlate this data by using the species averages and errors? 

Thank you,

Caio

My data consits out of observations rated on 5 different dimensions (5 point Lickert-scale). A sixth variable describes the outcome (binary 0-1) I am interested in. But rather understanding the individual contribution of these dimensions on the outcome variable, I am interested in finding out the optimal combination of dimensions resulting in the highest probability of getting the Outcome equal to 1.

Do you have any advice regarding a methodological approach for me?

Thank you very much in advance, your help is highly appreciated!

Kind regards,

Jessica Birkholz