Quantitative Research - Science topic

David: By "available data" I suppose you actually mean the entire population. Yes, to the point where the variance from the sample of n=N is zero.

A quantitative research will be relevant ( *but not exclusive* ) to highlight the many points of details making a difference in some linguistic aspect.

A guidebook:

Deborah Schiffrin, _Approaches to Discourse_ (1994).

Deborah Schiffrin et al.,

_The handbook of Discourse Analysis_ (2005).

(PDF) Immigration: analysis, trends and outlook on the global research activity (researchgate.net)

Hello Karismi,

Much of the answer depends on how you quantify scores for each of the ten circular use outcomes. As your question and Kelvin Jones' answer imply, the basic framework sounds like an anova (6 levels of IV motivation, 4 levels of IV values, 4 levels of proposed moderator, home ownership), with 96 cells and either 10 DVs or 10 univariate analyses or a repeated measures factor with 10 levels (scores). However, that requires your DV scores are rightly quantified as at least interval strength values.

If your attention is genuinely focused on the ensemble of 10 scores (and possible profile differences), then multivariate is likely the useful path. However, if your interest is really that of "which, if any, of these 10 scores is significantly influenced by IV1, IV2 & moderator," then univariate analyses, while repetitive, are likely the better option.

If there's something that we've missed here, please feel free to elaborate your query.

Good luck with your work.

In general, a questionnaire that relied on open-ended items would be considered qualitative, but I agree that it is important to emphasize things such as interpretation and meaning as purposes in qualitative research.

A variable usually implies something that has been measured, and which has variance. Neither of these apply to qualitative research in general, but if you are doing a form of content analysis where you count how often codes occur, then you are essentially treating those codes as variables.

Can you please send me your email?

Thank you Burhan Hadi for the excellent question and experts for the good answers.

This free manual provides an overall guide - it stress using a simulation approach to estimating power which is a very general one:

I don't know a paper about it, but we recently did a project with Kings Business School to address this quantitative skills gap.

The marketing students were challenged to optimise a digital shop to maximise sales for a product of their choice. Each group of students implemented different nudges within the simulated shop and create different adverts for participants to go through before the shopping simulation.

The top-performing student group increased sales for their chosen product by 3X, by a series of nudges, and adverts, compared to the control condition. It was really impressive work from the students, pulling together all the theories they had learnt to influence the 'shopper' experience.

The practical was supported by Gorilla Shop Builder, which makes it easy for students to do create this type of shopping simulation.

The lecturer and students are going to present their work at the BeOnline conference on Tuesday July 5th:

You might find it interesting!

Dear Benjamin Neumann . See the following useful RG links:

It is not unusual to discover links to the existing literature, once you have the results from an exploratory qualitative study. Unfortunately, there is no standard solution to how you should report this in either the qualitative or mixed methods fields.

I recommend to my students that when this occurs, they present it in the Discussion section. If your thesis supervisor is not willing to accept that as a compromise, then you will probably have to accept the power differential and do what you are being told to do -- for your thesis. Then, you will be free to do as you wish when you publish your research.

Yes and No! For a quantitative study, you are going to test relationships. It is widespread among scholars to state hypotheses, however, as long as you study the associations, you can dismiss the statement of the hypotheses directly (but you must have some sorts of research question or study purpose to see relationships) as long as you can prove these associations are significant or not. Stating hypotheses can help readers understand what you mean and prevent confusion.

Natasha Chironda The face is also measured in three segments: from the hairline on the forehead to the region between the eyes, from the area between the eyes to the bottom of the nose, and from the bottom of the nose the bottom of the chin. When these measures are measured, a person is deemed more lovely if the numbers are equal.

You may consider the TPB model or the UTAUT model explaining online purchase intention among students. Also, mediators should be considered, such as satisfaction or brand image. The COVID-19 pandemic could be considered as a moderator.

All the best,

Katrina Partol ,

The question about HR management in the hotel industry is very attractive to solve HR issues. Sometimes, ethical change management and talent management could be applied in this sector. Kindly visit the RG links.

All the best,

Yes, I support the responses of Sachin Suknunan and Rodney Duffett .

Kind Regards,

If you are interested in between place differences, your design will be very underpowered as you only have two places. If you want to ascertain whether place A is different on average from place B you will need lots of people in both places to draw conclusions with some certainty.

Jones, K. (1994). Analysing and designing extensive studies in which place matters: a comment on neighbourhood incivilities and the study of crime in place. Area, 79-81. https://www.jstor.org/stable/20003374

Reliability of a questionnaire means consistency of the results it renders. On the other hand, validity denotes the accuracy of the results’ inferences. A reliable and valid questionnaire is attained by ensuring the research concepts and models of selected questions/items are grounded in theory/experience, aligning with a study purpose.

A questionnaire’s reliability and validity can be assessed by examining its psychometric properties, encompassing reliability evidence (e.g., internal consistency) and validity evidence (e.g., content validity). It is recommended that you follow the psychometric evaluation standards established in your discipline. As an example, you might go through the standards by the Joint Committee on the Standards for Educational and Psychological Testing of AERA, APA, and NCME (2014), referenced below.

Joint Committee on the Standards for Educational and Psychological Testing of AERA, APA, and NCME. (2014). Standards for educational and psychological testing. American Educational Research Association. https://www.apa.org/science/programs/testing/standards#

Good luck,

I suggest to use t Test..in SPSS

Thank you Prof. Ines Carvalho then the way forward is to use the quotes as supplementing the quantitative results. Hope this helpful.

Dear Ma.Sherhan Punzalan Dolit

Purposive sampling is an appropriate type of sampling in particular circumstances. It picks instances based on expert opinion or with a specific goal in mind. Purposive sampling is most commonly utilized when measuring a difficult-to-reach population.

Purposive sampling is used by researchers when they wish to reach a certain group of people, as all survey participants are chosen because they meet a specific profile.

Best Regards

there is a lot of software for quantitative research. but this is dependent on you what do you want to do? and what level do you want to work at?

Generally, I think Eviews for Time series and Stata for Panel data for usual quantitative research are enough. but Python is better for more advanced quantitative research.

The four basic steps: Define, Appraise, , Synthesis ,and Analysis.

Of course, they should be added and used (qualitatively) -- it's data, and respondents felt a need to add them, so we should report them. Since the researcher did not ask for comments, it doesn't make sense to call it a mixed method study. It's a quant study with supplemental comments. You add a section to the report describing qualitatively, what was said, including areas of convergence and divergence if the sample is large enough, and with some representative verbatims in any case. You should also include the % of respondents who added comments, for context.

Check https://www.sagepub.com/sites/default/files/upm-binaries/22781_Chapter_3.pdf

secondary sources become important if the study contribute to previous literature. It all depends on your research questions or objectives.

As you are Spanish, I recommend classic manuals, understandable and unless it is Spanish (some, although they say as "Experimental Psychology", are valid for their objectives):

-Psicología Experimental; McGuigan; Ed. TRillas

-Diseño y Cálculo de Tests Estadísticos para Ensayos Clínicos y de Laboratorio; A. González; E.U.E de la Univ. Complutense de Madrid

-INvestigación Científica en CC de la Salud. Polit-Hungler. McGraw-Hill Interamericana

-Psicología Matemática I y II. Julio Seoane et al.Editorial UNED

-Fundamentos de Psicología Matemática.L. Jañez Escalada. Pirámide

-Psicología Matemática. Francis Isidro. Pirámide

-Los clásicos del Prof. Dr. Yela y los Clásicos del Prof. DR. Amón

Generally, Paid Journals are not of good quality Index. However, I am providing list of publishers publish both paid and free but are of good Indexing

1. Elsevier

2. Taylor and Francis

3. Springer

4. Hindawi

5. Willey

H1 and H2 could be examined by computing correlation coefficients and testing the null hypothesis that each correlation is zero in the population.

H3 (moderation) could be tested by using moderated regression where you regress Y on X and W and in addition include the interaction (product) term XW as a third predictor/independent variable to test the hypothesized moderation effect. (You can do the same for Z as an outcome variable.)

When the regression coefficient for the interaction term XW is significant, this means that there is a statistically significant moderation effect. To understand the meaning of the interaction, it useful to plot the relationship between X and Y at different levels of W. This is described in great detail with SPSS examples in Aiken & West's (1991) book on interactions in multiple regression. The procedure is somewhat complicated, also due to the fact that centering the predictors prior to forming the interaction term is often necessary for meaningful results. You therefore might want to study this topic in somewhat more detail before proceeding with your own analyses. You want to be sure that the results are correct and that they make sense and are easily interpretable.

Here I attach something about it:

-Social Perception of Gender Violence; © MINISTRY OF HEALTH, SOCIAL SERVICES AND EQUALITY. Publications Service. E-mail: publications@msssi.es;

htpp: // Wwww. official publications.boe.es.

In it you will find a lot of documentation about it.

-The Questionnaire of Attitudes towards Gender Equality (CAIG): elaboration and psychometric study, by Benlloch, IM. . Yearbook of Psychology; 2003. vol. 34, 1: 101-123;

htpps: //www.raco. cato.article. download.

I agree with David L Morgan as well. Fallibilism is an important epistemological stance here (which you also find in American pragmatism, like the philosophical work of Charles Sanders Peirce). Also think about Karl Popper's criticism of logical positivism and the falsification principle he proposed to be used instead of the verification principle advocated by logical positivism. If you would like to read more about post positivism as a philosophical basis for doing social research (in this case, educational research), I can recommend a very helpful book (but remember that the term "post positivism" can be interpreted and used in different ways, as Kelvyn Jones above mentions): https://psycnet.apa.org/record/2000-08253-000 . Good luck with your work!

Suggestion: Code your qualitative data into a small number of categories that reflect the main ways hotels cope. Assuming there are several hotels that fall in each category, you could do a one-way analysis of variance to see if the mean occupancy rates differ among the category groups.

One of maxims is proper prior planning Prevents Poor Performance. In this case if any number could be guessed what is the point of the question? Personally I don't think ANOVA on essentially random guesses is going to help. If the experiment is worth doing it is worth doing well. There are no Stati methods that can make bad data.good. My advice is go.to your advisor discuss it and if it's important redo the experiment. This happened to one of our students thanks.to last minute changes by a faculty member that destroyed her data. There was nothing to do but repeat it correctly.. Recall the above maxim. This is sometimes harsh but GIGO is another maxim in research. Best of luck, David Booth

Dear Hailemichael Kindie

without knowing your intended calculations, this is impossible to determine. I would refrain from using often unattainable "rules of thumb" like x participants per variable. You will not only find a number of threads on RG that have dealt with this question, but a mountain of literature on the topic of statistical power. One of the most precise, yet difficult procedures would be a Monte Carlo simulation - I am struggling with those myself at the moment.

Good luck!

Marcel

Som Nath Adhikari ,

Before getting into all the technical details, please consider whether it makes common sense. If you are looking at the influence of something like Facebook addiction on academic performance then you need some way of measuring performance, and performance should be your dependent variable. If you are only measuring this once then there is a serious potential problem with your research: you appear to be assuming that there is no relationship between Facebook addiction and prior ability, which I suspect is not the case. For a variety of social factors I would expect that smarter people are less addicted to Facebook and therefore appear to perform better.

Hi Sidra. I think we still can use the Kriejie Morgan Table of Sampling. In my field town planning research..many of the researchers still using this table. I'm graduated my PhD with minor correction in 2019 & I used this table to justify my sample size.

Research always is researching what we don't know. Thus, a research project always comes from a research question, a research problem. It does not come from any kind of previous "research design" which means the methodological or technical tools mobilized to address the research question or problem. Therefore, there are no easier or harder "research design" processes since they depend on your capacity to elaborate an original research problem. This is the most difficult. It is easier to replicate available research published through papers, books, etc., but this not scientific research but a copy and paste system that cannot be accepted in Ph.D. thesis, academic papers, or conference proceedings. Research design is a process as follow: 1. presenting the background of the research; 2. presenting the research question; 3. Statement of the problem; 4. theoretical and conceptual framework; 5. Significance of the research; 6. Scope and limitation of the research.

Yes, i agree. Behaviour, or in other words: i want to measure the attention participants had in the week(s) after the experiment. Did they follow the shown news, and where, after the experiment. To see if there is a difference in this behaviour for participants between the 2D and 360 condition.

Anyone a good suggestion?

Try to use online tools for research during pandemic.

A well-known downward-bias, due to partial observability of circumstances that affect individual outcome, and an upward bias, which depends on the econometric method used and the quality of the available data.

I cannot understand this point of 'old fashioned', any scientist does know that the attitude with which looks at the results has a deep influence on his/her conclusion. The example of PCA is only one of the infinite cases in which in my work I stepped into 'epistemological' crucial points. If I think philosophy has no relevance in science I will be condamened to adopt a bad phylosophy and not to grasp the meaning of what I am doing. It is the overlook of epistemology that generated the actual repeatability crisis of biomedical sciences. The data 'do not speak by themselves' but only when solicitated by a theory oriented perspective and any obtained model takes with it the strengths and limitations of an epistemological (even if often unconscious) perspective.

So said, we must not consider as epistemology the books of 'professional epistemologists' (that in many cases never stepped into a real laboratory) but the reflections of real concerned scientists about the sense of what they are doing.

I totally agree with the opinion of @Idarmis Knight Soto

To answer this question, let me repeat what I have written in my recording sheet during my bachelor study:

Now, the following points can help you to choose between the two methodologies:

Furthermore, there are other factors that have a great effect like cultural considerations.

Very useful discussion. I remember, the saying "it is better to study one rat for 1000 hrs rather than studying 1000 rats for one hr if you need to study behaviors of rats". But that particular rat should be selected carefully; not to choose a 'mad' or 'disabled' rat. Now by studying this carefully selected rat (theoretical sampling) we can generalize certain behavioral patterns of rats, as well as can gain a deep understanding of the behavior of rats, in that time and space. Observing 1000 rats, randomly or even purposively (statistical sampling), for one hour would help to identify a single or couple of features of rats, in that particular context. So, the sampling technique depends on the nature (not the purpose) of the research.

I am from cinema field, and the fact of proposing cultural heritage presentation throw fiction movies, made me discover during the whole four years of my study, every thing that is related to cultural heritage, either material or immatrial

what i want to say is YES! Yes you should discover another field and link it to yours, it may appear a bit stange in the beginning, but with research advance, every thing will be alright, knowledge evolution will take place on your mind, which will make your viewpoint to make your study more profound

as a conclusion, all the sience fields could be related to one another , the prodigies of history are in most cases multidisciplinar:

1-Avicenna was at the same time physician, astronomer, thinker, writer, of the Islamic Golden Age and father of early modern medicine

2-leonardo da vinci was at the same time a polymath, painter, draughtsman, engineer, scientist, theorist, sculptor, architect... and so on

........

if you will face a new field of science, linking it to your yet acquired knowledge, be sure that the result would be amazing ;)

Best luck

FRR

Rahmawati -

There is another approach, probably more often used for establishment surveys. The model-based approach. See for example Royall, R.M.(1992), "The model based (prediction) approach to finite population sampling theory," Institute of Mathematical Statistics Lecture Notes - Monograph Series, Volume 17, pp. 225-240.  Available under Project Euclid, open access:

 https://projecteuclid.org/euclid.lnms/1215458849. 

Here are a couple of textbooks:

Chambers, R, and Clark, R(2012), An Introduction to Model-Based Survey Sampling with Applications, Oxford Statistical Science Series

And

Valliant, R, Dorfman, A.H., and Royall, R.M.(2000), Finite Population Sampling and Inference: A Prediction Approach, Wiley Series in Probability and Statistics

The idea is that you use regression. So you have to have another data source for an independent variable on everything. That is the hard part in many cases, having the auxiliary data.

There are other books where the methods are "combined," as Ken Brewer has said, and he wrote this as well:

Brewer(2014), "Three controversies in the history of survey sampling," Survey Methodology, Dec 2013 -  Ken Brewer/Waksberg Award: 

When you can use regression, you have more choices for sampling because you have data on everything and know your population better. (You might look up "balanced sampling.") Sometimes auxiliary data are available from a recently past census. Sometimes there is administrative data available online for businesses, if you have that. But my guess is that you do not have the necessary auxiliary/regressor/predictor data available to you.

Best wishes - Jim

I prefer a research-specific book for my quantitative inquiry. Tips;

For General-purpose:

1. Research Methods for Business-A skill-building approach-Uma Sekaran & Roger Bougie (Wiley).

2. Business research methods-Alan Bryman & Emma Bell (Oxford University Press).

3. Designing and using organizational surveys-Allan H. Church and Janine Waclawski.

For data analysis purpose:

1. Discovering statistics using SPSS-Andy Field (Sage), you can also consult Julie Pallant's Step by Step SPSS.

Hope this would be of help.

Both are important for gaining different kinds of knowledge.

Robert Trevethan I disagree with you statement "that appending some open-ended questions to a set of fixed-choice questions on a survey in no way constitutes mixed-methods research."

In particular, if the qualitative content can be integrated with the quantitative content in ways that add to the overall understanding of the research question, then that is precisely the goal of mixed methods research. As I mentioned above, this is quite possible by using a QUAN --> qual explanatory sequential design in a survey that uses open-ended questions.

Ultimately, what matters most for mixed mentors is the purpose that the two types of data serve within the overall design, not when or how they were collected.

I agree with the others about starting with factor analysis.

With regard to industry-level and group-level, a variable can only be included in an analysis once. So, if you want to conduct separate analyses at each level, that is one thing, but if you want to include both in the same analysis, that will not work.

Basically, your model is y = x/y + other covariates. You can't have your dependent variable (revenue) both as dependent and independent variables.

unfortunately it is different in different countries, and in different universities. in our republic, for example, the more the better ((

A lot depends on your purpose and what you are trying to learn, as well as your field. Random/representative samples from a defined population are best, but they typically require a lot of resources/cost. In some fields most academic studies use nonrandom approaches which can be much less expensive. There are many ways to get such samples. I would look at papers in your domain to see the kinds of samples that people generally use.

Thank you very much.

If it is a literature review, it makes sense to organize data according to subjects and chronology. It is very common in historiography and other social sciences.

Ph. D basically in my opinion research oriented. Review or research article both research oriented but distinction innovation or invention. If any research or review innovation or invention oriented then eligible for ph. D. Next que some people says review not consider reason behind consider no lab work true but if nnovation then consider in ph. D because all bioinformatics chemoinformatics have no wet lab work that is similar as review..

But review must be innovation oriented.

Having two different data (Quan &Qan) deliberately does not mean necessarily is a mixed method, since, mixed method has its own research design criteria, which is taking aspects from positivism and interpretive positions as pragmatic paradigm. What is the reality and how do we know reality, are two different stories. Seemingly, the objective of paper is misleading the audience. Having quantitative results, following by qualitative procedure is not a mixed strategy, is only corroboration between quantitative results with qualitative findings, which paper resulted no difference. So, in my perspective the third paper is dealing with objective realities that exist, concerning human’s behavior and actions in corona virus crisis. As results falls into realist ontology with a post - positivism epidemiology.

To your first question, it is not right to look for quantitative data for inductive analysis. Inductive analysis is driven by reasoning based on the data emerging from observations and participants views. Inductive analysis is therefore based on qualitative data. Quantitative means you are either testing an hypothesis or a theory that you already know something about. My opinion.

Hi Behnam Abedin, you may consider using G*Power, a free power analysis program, to calculate your minimum sample size. Following is the reference article, download link and learning video. Hope it would help.

1. Reference article

Faul, F., Erdfelder, E., Buchner, A., & Lang, A. G. (2009). Statistical power analyses using G* Power 3.1: Tests for correlation and regression analyses. Behavior research methods, 41(4), 1149-1160.

2. Software Download (Windows and Mac)

3. YouTube: Using G*Power to Determine Sample Size

Thank you Mr.Morgan and Mr.Duffett!

And https://carlgans.org/report/example-of-quantitative-research-critique-paper/7/

I agree with Dennis Njung'e and Aruditya Jasrotia totally. Convenience sampling would do as well.

Dear Albandari Almutairi,

Your question has two (2) sides, relatively Quantitative Study and a PhD thesis.

If the findings of a study needs to be concretely supported with numerical findings from numerical data analyses, the study can be simply categorised as a quantitative study.

Generally, information can be in words and numerically transformed data. when a study has focused on numerical data to support its arguments, confirm hypothesis and explore the findings, such a study can be a quantitative study. On the other hand, if the study has focused on qualitative information to support its arguments, confirm hypothesis and explore the findings, such a study can be a qualitative study.

Whether the study can be considered for a PhD progression, it depends on the depth and quality of the investigation made in the study. If your quantitative investigation has sufficient depth of analyses and conclusion with well-designed research methods and approaches to explore new findings or methodology, such a study can be considered for a PhD dissertation purpose, subject to an authentication by acceptable scholars.

Hope this simple explanation can help you. Good luck in your studies.

Regards,

S. Senthilnathan.

Dear Prof. Igor Gurkov,

Thanks for your guidelines and answers. Much appreciated your time and insights!

We live in an age of quantification for profit.

Putting the economic interests aside, quantifying is about mass data, and qualitative data is about personal data. Predicting human behavior, in both cases, is about manipulating behavior to buy into something. As you also mentioned data (normal distribution), we should avoid to becoming captured by data.

Here, we arrive, at the moment of statistical learning theory.

Daniel Wright I will do research into the hotel industry in general, located in the Netherlands. I do think I will use stratisfied sampling for my online questionnaire to get an equal result from female and male and purposive sampling for my interviews with hotel managers. My sampling size is 382 respondents (95% confidential rate, 5% margin error. I am not familiar with non-respondents.. will take a look into this.

Thank you Cristian Ramos-Vera, your information was so valued for carrying my research project.

Shamas Suleman Arshad, I'm sorry, but I don't understand your question, and I'm not sure that the above posts address it for you. I wonder whether you would consider rewriting your question a little bit so that it might be easier to help you.

Apart from that I hope it's OK if I point out that using the Kaiser criterion (eigenvalues > 1) to identify the number of factors in a dataset has been criticised for at least 30 years. For some reason, however, the Kaiser criterion keeps getting used (do researchers simply follow each other like sheep?) despite the scree test, or, better, parallel analysis being much more helpful.

In my experience, the Kaiser criterion often suggests there are more items in the data than is reasonable / interpretable, and the scree test is sometimes too ambiguous. As a result, I much prefer to use and recommend parallel analysis - particularly because it's very easy to perform.

Good point David

Thank you Glenn Wayne Jones .

I think I can proceed from what you have mentioned above.

May i know if there is any papers/articles which have done the process, if you can suggest a few?

Thank you

Shamsinar

Can also refer to the article below :-

Golafshani, N. (2003). Understanding reliability and validity in qualitative research. The qualitative report, 8(4), 597-607.

If I understand your planned analysis correctly, there will be two main steps in the analysis, namely, the formation of the codes/categories (commonly dummy variables that indicate the presence of a particular idea or theme) that will be used in the content/theme analysis, and secondly the coding of each participant's response against each of these codes. It will be difficult to determine in any systematic manner the level of agreement between coders unless they are using the same set of coding categories.

To derive a commonly agreed upon set of coding categories, a common approach is for coders to use a randomly selected set of responses to separately form a set of coding categories, that are determined, not only by the responses, but also by the purpose of the research or any specific hyp