Two Way - Science topic

If you start writing short pieces that do not need full peer review it gives good experience in dealing with editors and reviewers. They can sometimes be very difficult to work with.

Book chapters are a good start. It is usually easy to find people seeking research collaborators for books. Expertise is more important than a past publication record.

You can of course seek collaborators yourself by post an inquiry on RG for instance. Lots of people do.

This is a deeply important question, and it lies at the heart of what my work — Quantum Wavefront Dynamics (QWD) — was created to address.

The short answer is:

No, wave-particle duality is not caused by wavefunction collapse.

Collapse is not what creates the duality — it’s what destroys the wave-like behavior and reveals the particle outcome.

But the mystery goes deeper:

How can one entity behave like both a wave and a particle, and when does it become one or the other?

According to QWD:

• The quantum entity (photon, electron, etc.) is always a real, extended wave — not a probabilistic ghost.

• This wave has a geometric structure — a wavefront — that propagates deterministically.

• When no external disturbance occurs, this wave evolves coherently and produces interference patterns.

• However, when the environment (observer, device, material) interacts with it asymmetrically, it begins to distort the wavefront curvature.

• This distortion leads to gradual localization — the wavefront bends, concentrates, and behaves as if it “collapses” into a particle detection.

QWD’s Key Contribution:

• It provides a single physical description — not duality, not collapse postulates.

• Wave and particle are two stages of one evolving wavefront under different physical conditions.

• Collapse is not instantaneous. It’s a continuous, deterministic, geometry-driven process.

This model aligns with modern experimental findings (like decoherence and interference decay) but removes the need for mystery, randomness, or parallel realities.

If this resonates with you, I invite you — and all reading — to explore my published manuscript:

Quantum Wavefront Dynamics (QWD): A Deterministic Framework for the Wave-to-Particle Transition

Available here on ResearchGate: [Insert Your RG DOI Link]

It presents a full theoretical explanation, simulations, and proposed experimental tests — and offers a unified view that dissolves the duality paradox.

Sincerely,

Adam Dakhil Nasser Alzerkany

Author of QWD

Yes, you can run one-way PERMANOVAs as post-hoc tests if properly corrected for multiple comparisons. Consider pairwise PERMANOVA (if available) for a more direct approach. Check assumptions (e.g., dispersion homogeneity via PERMDISP) to ensure significant results are due to mean differences, not variance differences.

Dear Mohammad Zarandi ,

The answer to all three questions is positive.

The second and third questions are related. Which foregrounds the perspective of direct marketing... Since it is very important that the customer trusts the bank, this is also true in the case of the bank.

The questions and the answers given can also be examined from a cause-and-effect perspective.

Regards,

Laszlo

SQ: <<So I think it is the case to understand what are the conclusions of SR about the result of the proposed configuration.>>

Yes. Anybody interested has got the opportunity to read my former posts.

Eg. #9 on p. 461 [my first of July 3, 2024] within

https://www.researchgate.net/post/Is_spacetime_an_elastic_medium_that_propagates_waves?

proves reciprocity of time delay in SRT.

The scenario of the preprint is analysed in MANY OTHER of my posts.

If you obtain an overall p value less than 0.05 for the total ANOVA, you then do pairwise comparisons between the variables to find out which have significant interactions. You don't need to do anything with slope values :)

You have detected differences in the groups and a change in the outcome over time, but neither of these effects is related to the intervention. In a pre-post design the interaction is a test of the intervention because you'd expect the pre intervention scores to be similar (especially in a randomized design) and so the intervention effect should show up in the post scores only. This effect (the group difference on post minus the group difference on pre) is the intervention effect and what is tested by time*intervention.

One puzzle is that the groups differ at pre as well as post which suggests its not a randomized design (or you were unlucky).

Dear Doctor

"Method to overcome the problemThe vanishing gradient problem is caused by the derivative of the activation function used to create the neural network. The simplest solution to the problem is to replace the activation function of the network. Instead of sigmoid, use an activation function such as ReLU."

Hi Divit Karnawat.

I'm simulating a water injection into a tank, and currently, I have the same problem. I parametrized the velocity and mass flow as a profile based on a transient table, but the temperature doesn't show this option. Did you find a solution? Please, I hope your answer.

Best Regards

I am also interested how to perform a two-way ANOSIM in R (ideally using vegan package). Any reply to this?

So far, I did it using PAST software (https://www.nhm.uio.no/english/research/resources/past/) but I would appreciate an R-code.

Use the emmeans or lsmeans option in the software you are using. This is a post-hoc test after the anova, and accounts for the repeated-measures nature of the model. Often you can get confidence intervals for the e.m. means. Plotting or reporting these intervals with the e.m. means is useful for the reader.

I think that the classical peer-review/publishing approach only makes sense if it is extremely rigorous on every aspect of the review. That is, peer-review only makes sense if it actually is a quality guarantee, like with an ISO standard to follow, but time has shown that it is very far from it. I'm not even considering predatory journals and the like, as even top journals have shown to be prone to publish the most ludicrous stuff, with the most obvious problem being that reviewers don't really know anymore anything at all these days (so a lot of obvious or old stuff gets published again, not to mention completely wrong stuff). Obviously, there might be differences between fields, but it is especially so in STEM.

One should also take in mind that research and the relative publishing (should) have nothing to do with employment and career paths, at least officially. So, when we think about it, we should not, honestly, try to fix it in order to make it work as a job market tool. I hold this truth to be self evident.

So, the relevant question is, in my opinion: how do we save research and publishing from the misuses that make it today so low quality? My answer is that, unfortunately, we can't, or at least not yet as of today. We currently live in a society that has had a repentine shift in a lot of paradigms, with the result that we failed in preserving the cultural background of most aspects of our life because of the speed. So, the current society, not anymore aware of the past, needs to understand from failure what could have been preserved by culture.

I think that a suitable model could be just like a wikipedia, with each article being an independent and dynamic entity, subject to fully tracked non blind reviews from whoever thinks to do so and the relative changes exposed. Most importantly, nobody can impose anything on the original author. But this will probably never happen in the near future.

ANOVA (Analysis of Variance) is a statistical test used to analyze the difference between the means of more than two groups. A one-way ANOVA uses one independent variable, while a two-way ANOVA uses two independent variables.

Dear Alok, thanks a lot for your answer.

I read an article on arXiv that proposes a new estimator for the average causal effects of a binary treatment with panel data in settings with general treatment patterns.

Hello Bieke,

If you genuinely believe that the unit difference separating "no frame" and "low" is exactly the same as that separating "low" and "high" for an IV, then yes, you may code them as -1, 0, 1 (or any other linear transformation thereof). However, if you don't believe this, then you must opt for recoding, replacing a three-level variable with two dummy variates.

I suspect that the reason one of your variables is being dropped is due to redundancy, and this could be caused by: (a) no instances of one of the three categories in the data set; (b) IV1 and IV2 are redundant in some way (e.g., every case is in the same level for both variables); (c) a mistaken recoding.

Are you also including all four of the dummy variate products to capture the information about their interaction?

Good luck with your work.

Hello Isabel,

The basic design is a two-between (intensity, 2 levels; protein, 2 levels), one-within (occasion, 2 levels) manova. Between-within manovas are sometimes called "doubly multivariate."

So yes, you have three factors.

However, you may wish to consider a couple of points:

1. Do you really intend to interpret the results across arbitrary linear composite/s of the five DVs (generated by the software to maximize the variance accounted for by the between-subjects IVs), or would you prefer to address how the study factors influence scores on each of the five DVs? The first calls for the multivariate test; the second calls for univariate tests.

2. Consider using the pre-intervention score for a given dependent variable as a covariate, then use the post-intervention score as the DV score. This eliminates the within-subjects factor, and offers the ability to "adjust" for randomly occurring differences at the outset among the four treatment combinations.

3. In general, tests of change over time (e.g., meanT2 - meanT1) force you to work with difference scores that are less reliable than either of the scores from which they are derived, unless all time scores are perfectly reliable (no measurement error), which is seldom if ever the case.

Good luck with your work.

This error can have several reasons. One of the reasons is applying sudden pressure from the fluid to the solid part in the initial time steps. Defining an extra stiffness at the initial time steps (i.e., ten times higher) can be helpful to end this issue.

This additional stiffness help to prevent distortion due to the suddenly increased force in the interface of solid and fluid parts, and, after 1-5 time steps, disappear and reach to normal stiffness of the materials.

Shuo Yuan , from what I recall the version of the Auricchio model that is available in ANSYS cannot be used to simulate the two-way shape memory effect. In order to do that you could either write your own material sub-routine or try to mimic this effect by using different coefficients of thermal expansion for each phase (the latter is suggested only for investigation purposes and should be considered as a simplistic and approximating approach).

Best regards,

Anargyros

Hi

Have your reached any conclusion? Currently, I am experiencing similar difficulty.

I have the same question! although may manuals and websites state that it is robust there don't seem to be any peer reviewed references for two- or three-way ANOVA. I can find a couple of references for one-way and RM ANOVA (PMID 36695847 & 29048317). How would you justify carrying out an ANOVA on non-normal data in a paper/thesis?

Yes, you should include your control group when doing any analysis because you will want to know if there is a difference between the control group and the experimental conditions. For example, you would have no way of knowing if having a bare hand is different than wearing an oversized glove. By including your control group it allows you to statistically determine if there is (or is not) a difference.

You could still use a two-way ANOVA with your questions, though. Here is what I would do (use your judgment for your own research)

1). Glove fit + surgeon status as predictor of sensitivity

This will give you a comparison of is a surgeon with no gloves has different sensitivity compare to non-surgeons who have over-fitted gloves, etc.

2). Will be taken care of in the above analysis. When you run your two-way ANOVA, be sure your analysis will report interaction effects and multiple comparisons.

3). This will also be taken care of in the first analysis.

A two-way ANOVA with interaction effects and multiple comparisons will report if there is a difference between surgeons and non-surgeons' sensitivity while wearing gloves of different thicknesses.

I hope this helps.

It is preferable to present both wt% and at%. How you use the data depends on what you want to show. If you want to identify the presence of a compound, the use of at% will allow you to obtain the chemical formula. If you are checking elemental levels in an alloy, then wt% is used.

Attachment:

M1 is the smallest particle, which is the smallest theoretically and experimentally: indivisible: m1 ≠ mx+my.

The smallest particle ≠ the basic particle in the experiment.

The smallest particle m1 is proved by "theory and reality", that is, the universe must contain the smallest particle m1.

Hello Hannah Bauer. Given that your discipline appears to be psychology, and given that you are reading Andy Field's book and looking at a website with spss in the address, I'll guess that you might be using SPSS as your primary stats package. If so, have you considered using the MIXED command rather than GLM? MIXED has options for allowing/modeling heterogeneity of variance. Take a look at the examples on these two UCLA pages.

HTH.

Better to replant the native Pseudomonas bacteria host plants, as those bacteria are the best cloud seeders on the planet. Image of a cloud being born here in California. Read article at https://www.discovermagazine.com/planet-earth/does-rain-come-from-life-in-the-clouds

Hi,

Software Gpower has modules on ANOVA.It is free to use.

Here are some good books on this subject:

Meinhard Kieser Springer Series in Pharmaceutical Statistics Methods and Applications of Sample Size Calculation and Recalculation in Clinical Trials Springer2020

Rens van de Schoot European Association of Methodology Small Sample Size Solutions: A Guide for Applied Researchers and Practitioners Routledge2020

J. P. Verma, Priyam Verma Determining Sample Size and Power in Research Studies: A Manual for Researchers [1st ed.] Springer Singapore; Springer2020

Shein-Chung Chow, Jun Shao, Hansheng Wang, Yuliya LokhnyginaChapman & Hall/CRC biostatistics series Sample Size Calculations in Clinical Research, Third Edition [3 ed.] CRC Press; Chapman and Hall/CRC 2017

Tania Fernandes You want to see effects for all two-way interactions in the SPSS output, is that correct? If so, you want to do the following:

1. Analyze > General Linear Model > Univariate

2. Enter all independent variables and dependent variables

3. Select "EM Means"

4. Move all the items listed in the "Factor(s) and Factor Interactions" box over into the "Display Means for" box on the right

5. Check the box next to "Compare Means"

6. Below "Confidence Interval Adjustment" select the "Bonferroni" option

7. Click Continue

8. Click OK

Daniel Wright , I would bet the reason is that no such "large step" is seen here. Texbooks are full of stupid sentences like "the WMW-test is the non-parametric alternative of the t-test" and "if you want to do a t-test but your data is not normal distributed, then you can use the WMW-test instead", or even: "if you want to compare two samples and you cannot ensure that the data are normal distributed, use the WMW-test" and so on. This has left behind its traces...

Helllo,

please check the book Rand Wilcox (2017) - Introduction to Robust Estimation and Hypothesis Testing, chapter 8.6 Between-by-Within and Within-by-Within Designs.

Best,

Rainer

Measurement refers to the assignment of numbers in a meaningful way to variables. We have four scales of measurement: Nominal, ordinal, ordinal and ratio scales. The choice of scale depends on what you are looking for.

Yes, you have sampled both levels of each factor (sediment sandy/muddy, mangrove presence/absence) for both levels of the other, so it is a crossed design.

ANOSIM doesn't require a balanced design (though imbalance may lead to a loss of power).

Comparing the R values in a 2-way crossed ANOSIM will allow you to assess the relative strengths of the effects of sediment (removing effect of mangroves) vs the effect of mangrove (removing the effect of sediment).

Dear friend.

I recommend: https://books.google.com/books/about/Linear_Models_and_Time_Series_Analysis.html?id=vkNxDwAAQBAJ&redir_esc=y

He will answer all the questions :)

Usually when you have a significant interaction (e.g. Region x Gender), it is the effect of this interaction that is of interest, and the main effects (e.g. Region) are less of interest, because, in this case, we know that the effect of Region changes across levels of Gender.

An interaction plot for the two-way interaction often tells the story for the reader. ( e.g. https://www.researchgate.net/profile/Arild-Hestvik/publication/356982099/figure/fig3/AS:1100320079908864@1639348158015/Interaction-plot-for-the-2x2-design-GROUP-x-Condition-interaction-in-TF04SF1-Error-bars_W640.jpg )

Go to YouTube. You will find many examples of using two way Anova.

If it is a copyrighted work, you have to seek permission from the owner of its copyright before you publish it in your journal. Otherwise your action will fall in the infringement of the copyright.

Hamed Nooraldin

In general, when a curve is concave down, the trapezoidal rule will underestimate the area because when you connect the trapezoid's left and right sides to the curve, and then connect those two points to make the trapezoid's top, you'll be left with a little space above the trapezoid.

The trapezoidal rule employs function values at nodes that are equally spaced. It is extremely precise for integrals over periodic intervals but is frequently wrong in nonperiodic circumstances.

The Trapezoidal Rule is the average of the left and right sums, and it typically provides a better estimate than either alone. Simpson's Rule approximates area by using intervals topped with parabolas; hence, it delivers the precise area beneath quadratic functions.

However, if the integrand is concave up (i.e. has a positive second derivative), the error is negative and the trapezoidal method overestimates the real value.

I suggest to see the following tutorials as well:

1. https://www.youtube.com/watch?v=ud3RQymhk0g

2. https://www.youtube.com/watch?v=MQx3Bo1uurM

3. https://www.youtube.com/watch?v=cz30Jaqtghk

Thanks so much dear professor

Dear Deborah Lagura Herrera,

I suggest you to see links and atttached file on topic.

Best regards

Dear Anna!

Please you look at the protocol:

Two-way MLC

Two-way MLC were set up using various types of cells, i.e. whole blood mononuclear cells, CD2+ cells, CD2+CD16− cells, CD2+ CD4− cells, CD2+ CD8− cells, or cells preactivated by OKT3 stimulation. A total number of 2 × 107 cells was cultivated in cell culture flasks in 20 ml of culture medium with or without 20 U/ml IL-2. Medium was exchanged when its colour changed to yellow, usually once or twice per week. Initial cell ratios between the two populations of allogeneic cells were variable, as indicated in the respective experiments. Cell counting error in the 50:50 setup was < 3%. Cell cultures were analysed for composition of cell populations and subsets at various time points using 21 days as end point.

For proliferation assays and determination of cell numbers after different periods of culture, 2 × 105 cells/well were cultured in 200 μl culture medium in triplicates using 96-well U-shaped microtitre plates. In the one-way MLC, 105 cells, ‘responder’ cells, were mixed with 105 irradiated (30 Gy) allogeneic ‘stimulator’ cells; in the two-way MLC, 105 cells of each of the two allogeneic populations were mixed. Following 2, 4, 6, 8, 10 and 14 days of culture 3H-thymidine (0.5 μCi/well, 5 Ci/mmol; NEN, Dreieich, Germany) was added and plates were incubated for 8 h. Cells were then harvested with an automatic cell harvester (Pharmacia LKB, Uppsala, Sweden). Activity was measured in a liquid scintillation counter (LKB Wallac, Turku, Finland). The data are given as mean ct/min of triplicates ± s.d.

Mixed lymphocyte reaction Patient and healthy control peripheral blood mononuclear cells (PBMCs) were isolated using Ficoll-Paque PLUS (GE Healthcare Life Sciences, Little Chalfont, Buckinghamshire, UK). Responder and stimulator cells were labeled with proliferation dyes (CellTrace CFSE and CellTrace FarRed, respectively) according to manufacturer protocol (Thermo Fisher Scientific, Waltham, MA, USA). Isolated PBMCs were stained with their corresponding proliferation dye at a 1:1000 dilution and incubated in the dark at 37˚C for 20 min. After incubation, PBMCs were washed in culture media (Roswell Park Memorial Institute [RPMI] 1640 medium; Mediatech Inc., Manassas, VA, USA) containing 10% fetal bovine serum (FBS; Thermo Fisher Scientific) at five-fold volume of the volume used for staining and incubated in the dark at 37˚C for another 5 min. Stained PBMCs were centrifuged and resuspended in culture media (RPMI containing 10% FBS) at a volume resulting in a final concentration of 1 £ 10⁶ per mL. In the one-way MLR, stimulator cells were irradiated at 25 Gy. Responder and stimulator cells were co-cultured in 96-well plates at a concentration of 1 £ 105 cells per 100 mL each in RPMI media containing 10% FBS for 6
12 days. On days 6 and 12, 100 mL of the responder/stimulator mixture was removed for analysis using flow cytometry. In the two-way MLR, neither cell population was irradiated [15]. In two-way MLRs using normal healthy controls, both cell populations were plated at a concentration of 1 £ 10⁶ per mL for a total of 2 £ 10⁵ cells per well in 96-well plates unless otherwise noted. In two-way MLRs using patient samples, patient PBMCs were labeled with a single proliferation dye (CellTrace CFSE) and plated at 3 £ 10⁵ cells per well (no addition of donor and recipient cells). The patient PBMC amounts per well were not adjusted by donor chimerism in each patient. In parallel, PBMCs from a single normal healthy control (without mixture of another control cells) were also cultured to evaluate baseline lymphocyte proliferation in the same condition. A unique healthy control without cryopreservation was used in two-way MLR for patient samples, who was not the donor used in the original transplant. Flow cytometry For surface marker analysis, isolated PBMCs were labeled with antibodies against CD33 (clone AC104.3E3, Miltenyi Biotec, Bergisch Gladback, Germany), CD20 (clone L27, Becton Dickson, East Rutherford, NJ, USA), CD14 (clone TUK4, Miltenyi Biotec), CD8 (clone BW135/80, Miltenyi Biotec) or CD4 (clone M-T466, Miltenyi Biotec) for 30 min in the dark at 4˚C. A non-stained control was used to differentiate between positively labeled cells. Analysis was performed using a FACSCalibur flow cytometer (Becton Dickinson). For analysis of lymphocyte proliferation, a forward-scatter/side-scatter plot was used to gate the lymphocyte population. Non-proliferating and proliferating cells were captured in separate gates, and the percentages within each region was estimated (supplementary Figure 1). The proportion of living and dying cells was determined using the percentage of each cell population captured in either the combination of gates 1 and 2 or gate 3 (supplementary Figure 1) after 12 days in MLR culture. (attached file)

In my lab we use the Jeremy-Dawson excel to plot the three-way interactions. It is an excel template that you only have to introduce the required data.

Link to web: http://www.jeremydawson.co.uk/slopes.htm

Direct link to excel (linear model): http://www.jeremydawson.co.uk/3-way_linear_interactions.xls

As a general rule (valid for N > 30), the statistical error halves, if the sample size is increased 4fold. I would say, that in randomization tests you get a rough estimate with 250, a reasonable result with 1000 and you are on the safe side with 4000. More than 10000 is rarely necessary. As the default number of permutations of anosim in in R is 999 (+1, the observed case), you don't make anything wrong with 4999.

In a balanced full factorial design all effects are independent of each other (orthogonal). So the sums of squares attributed to two-way interaction can in theory and practice be 0 and the three-way interaction have non-zero SS and be significant.

Its unusual in practice because in general higher order effects account for less variation than lower order effects. This is the hierarchical ordering principle. Note this is a general tendency that arises for a mix of different reasons - it isn't inevitable:

This is a good example of why it's important to not place too much importance on p-values.

The significant p-value tells you that the statistical procedure is able to identify an effect of the interaction against the noise of the variability of the data * . Looking at the plot, your eyes, too, tell you that the slopes are different relative to the variability in the data. The error bars on the red points overlap, and those on the blue points do not.

But this significant result does not tell you that the interaction effect is large, nor that it is of any practical importance. The plot is very helpful for the reader to understand the results. The difference in slopes is small relative to the difference in Accents, and probably relative to the difference in Primes. How you interpret these observations for your reader is up to you.

__________

* This description is not technically correct, but hopefully gives you a sense of the point I'm trying to make.

please follow the link below

To clarify Alan F Rawle's answer:

Particle Size - the physical size of the particle

Crystallite Size - size of a single domain with the same orientation

Therefore, particles typically consist of multiple crystallites.

You are inherently measuring the different definitions above by using different techniques, and therefore MUST differentiate between them. TEM, specifically STEM (s is for scanning), can get you overall particle size. Diffraction within the TEM can give you lattice parameters and in turn the unit cell, however, this is fairly time consuming, complicated, and out of the scope of your question. XRD, by whichever analysis technique (Scherrer, integral breadth, Rietveld, etc.), you are measuring the scattering domain size, i.e. crystallite size.

The condition when particle size = crystallite size is when the particles are single crystals. In my personal experience, this condition rarely occurs. I've measured multiple nanoparticle systems on substrates and compared with XRD, and "agreeable" values are normally 10-30nm different. The reason for this deviation is the inherent broadening of nanoscaled materials in the XRD pattern. Also, in any type of microscopy, there is a limited sample size of particles, therefore counting statistics are poor.

Hope this clears some concepts up!

You will be best judge I think . There are so many model papers of logistic regression on Michales Menten equation - https://www.youtube.com/watch?v=fv6OAVe2vQc

I changed my time step to 18 and it worked. Thankyou!!

First David Morse is correct in everything he said. I would like to add that robustness of nonparametric tests to similar parametric tests is oftentimes just not true. You might want to look at the attached google search for some studies on the topic. Best wishes, David Booth

Mathematically (statistically), it is possible to present interaction effect in absence of the primary effects. Using effect size plots (countered plots) is a good assistant n this case.

Thank you so much Andrew Paul McKenzie Pegman !

Deepak Kumar Pandey Thank you Sir for your information

you should found a summary () of data before and use the lm and plm functions a

Thank you for your time and help Dr. Jos Feys

I've just opened the data and run a series of simulations with different means, standard deviations and sample sizes through rnorm() function. Now, I understand the difference better. As far as I understand, we can say that an ANOVA is a kind of analysis depending on a linear regression. I mean an ANOVA is a special case of GLM. That's all. Thank you for sharing it. It helps a lot.

Harish Kamath , Do you mean one-way FSI? As far as I know, steady-state analysis is defined as one-way FSI, and system coupling is no longer needed. In these simulations convergence is controlled by pressure-velocity coupling scheme and URF.

Hello Hosna,

Can you furnish screen shots of: data view and variable view for your data set? Be sure that the data view shot includes a couple of cases from each level of your religion variable. As well, try running a crosstabs on the two IVs that you're using, and see if that correctly separates the religion levels in your data set.

Good luck with your work.

Thank you Iyo

Unfortunately you are repeating what you said before. Therefore, for me at least, I am still in the dark

One more: What are entities?

Categorization of your continous variable would mean reducing information that lies in this variable.

When you are looking for an alternative that can deal with categorical as well as continuous IVs, you should consider Generalized Linear (mixed) Models.

Thank you so much!

In an ANCOVA test the covariate does not vary among the groups of the categorical factor or factors. Assuming that the interaction term is not significant then the assumption is not violated. If the interaction were significant,then you need to use a different approach to analyze this data, such as a mixed model.

Hi Rachel

I think you should use a 2-way MANOVA not 1-way MANOVA.

The following is an incredible resource:

http://www.statsmakemecry.com/smmctheblog/stats-soup-anova-ancova-manova-mancova#:~:text=Oneway%20ANOVA%20has%20one,of%20Education%20and%20Zodiac%20Sign).

In your measurement model, if the latent variable is indicated by both categorical variables and continuous variables. How do you proceed to choose one of the categorical variable fix to 1, or one of the continuous variables?

There is no rule. Choosing a continuous variable may make the interpretation simpler.

If you are trying to fit a one-factor CFA model in MPlus with 1 continuous indicator (values 0-10, and skewed distribution) andlets say ,10 categorical variables (4-option Likert), you may employ ULSMV. You may want to combine continuous and categorical indicators with ULSMV.

In attachment an R script for your design with the post hoc tests.

Yes, do run a between x within repeated measures ANOVA. In attachments R and SAS programs

I don't see a reason why you could drop the sphericity assumption. Greenhouse-Geisser is a typical solution, and some authors recommend to use it anyways, because you cannot rely on the results from Mauchly test (e.g. Maxwell & Delaney,

2004).

Other alternatives could be an MANOVA approach to repeated measures (Tabachnick & Fidell, 2007) or multilevel modeling. Both have pros and cons, but the latter is apparently the most flexible tool to incorporate different variance-covariance structures.

Assuming you wish to maintain good control over the family-wise alpha, Fisher's LSD should only be used when you have k=3 groups. Only in that case, does it control the family-wise alpha at the per-contrast alpha. David Howell discussed this in his textbook, Statistical Methods for Psychology (many editions available). I believe Thom Baguley also discusses it in his book, Serious Stats. And the following article is another source, if you need citations to convince a supervisor, co-author or reviewer (etc.). HTH.

PS- Fisher's LSD also assumes reasonably good homogeneity of variance, IIRC.

Aditya Chakraborty, you can consider this R package for RSA at https://www.nicebread.de/software/RSA/index.html

José Augusto Solís Benites sé Augusto Solís Benites ,Do you have a repeated measure or ordinal/categorized time data? How many IVs(15?), and type of DV. Interaction between what? To me, the tests you are using is inappropriate, misspecification exist, in your model(data.aov) 1. The Friedman test is the non-parametric alternative to the one-way ANOVA with repeated measures. It is used to test for differences between groups when the dependent variable being measured is ordinal. 2. The Jonckheere-Terpstra test is a rank-based nonparametric test that can be used to determine if there is a statistically significant trend between an ordinal independent variable and a continuous or ordinal dependent variable.(if the peruvian fish characteristics can be put in ordered scale.) Except we see the data, know the IV and DV structure, you are the one that knows best what to do, any advice here is off-mark or serious error(meaniless) given your unknown research questions and dataset structure.

Muhammad Ali Edward Peake Thanks for sharing. I will check them out.

Please go to this link:

https://www.timetech.de/products/new-trends-in-two-way-time-and-frequency-transfer-viasatellite/

If one drill well with drilling fluid with mud weight under balanced then there are chances of formation damage due to imbalances of hydro static pressure and formation pressure. Caving may occur which is if not reduced may lead to stuck of drill strings.

Sportscode, Dartfish.

Hello Glauco,

A search of IBM SPSS 24 Algorithms outlines the approximate SE method used to evaluate Tau-b in the Crosstabs subprogram, but no mention is made of what goes on in the Correlations subprogram.

I tried a sample data set with each in SPSS 26: Both yielded the same Tau-b value, but differed in the computed probability under the null hypothesis. Therefore, it is safe to conclude that these two subprograms evaluate significance level differently (and were likely implemented by different developers). Caveat emptor!

Good luck with your work!

Hello,Larry

You can use general lineaer mixed model(GLMM) to estimate any within and between effects.

GLMM id more general approach than repeated measures ANOVA .