Statistics, in my opinion, is a skill-set to be learned as the need arises.  It may first arise when reading professional papers.  It may next arise when you do your own study, as a means of comparison.  So much can be done on line, or with the help of computerized statistical packages.  Yet, it still comes down to:  what do I need to know, and how can statistics aide me in my work.

Ms. Davenport, 

Please see: https://www.pearson.com/us/

The school I worked at used this website for all their basic statistics classes.

Hope this helps.

I find that if students are asked to complete a roundtable workbook relevant to only their own research project before we have lab classes, then they fully engage with that and they have a framework in mind for their data entry, organisation, and hypothesis testing. You can access a quantitative-qualitative workbook that I wrote for my theses students on SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2407141

Hi Carole 

what has worked for me was to engage teams of students in market research projects. They worked on practical issues that were of great interest for them. The gathering and analysis of data was then of high relevance for them and they were pretty eager when dealing with statistics. 

I think that e-learning is a great deal for many universities, although I prefer a personal setting. However, I think that what you mentioned a 24/7 online-support would be a great service to offer to students. Best of luck for this project!

Ms. Davenport,

An e-learning platform like Moodle enables the implementation of teaching scenarios.

Students can download resources (Piece of paper) or links to videos,....but learning takes place when you ask the students to solve a problem.

Assignment can be requested individually or in groups.

The platform allows you to allocate them a collaborative editorial space (Wiki), a videoconference for synchronous discussions and many other tools.

A forum dedicated to each activity can be very useful: students can ask questions when they are stuck and are encouraged to respond to their peers when they think they have the answer. This reduces frustration and reinforces the sense of belonging to a learning community.

Thank you, I will look into Moodle. It may not work for the support centre as this is not a set module, but may be a good resource for lecturers teaching the statistics modules to look into.

This website shows from basic to advance with tutorials that are not too difficult to follow, even for beginners.

Best regards,

Debra

http://stattrek.com/

Thank you to everyone, there are some really great ideas here and I will be looking into them all. Keep them coming, there can never be too much in the way of resources as we all know that different student learn in different ways!

Methods of teaching statistics to higher education students

Not easy to teach students mastering statistics especially those who are undertaking their research.  Reason being scope of statistics are very wide e.g. non- / parametric statistics, analyses of variance, correlational studies, multiple regression, structural equation modeling etc. in which a student's actual research might only need e.g. 1 or 2 statistical analysis whereas the rest they might forget after a while not practicing them.

One suggestion to tackle above issue is to let the students re-attend their repetitive statistical classes numerous times even though they have attended before in the previous semesters.  This can allow them to focus & master those statistics that are relevant to their current research, the rest they can learn slowly & steadily in other future semesters.

I am voting up to my all my colleagues above and especially Debra Sharon Ferdinand-James and Han Ping Fung. I do agree with their ideas.

Emad

Excellent ideas.Anything new in addition to these?

The issue with that though is the availability of teaching resources and classroom sizes. For lower level degrees, statistics lectures usually have upwards of 150 students in the class and are then split into smaller groups for workshops, so it would not be possible, given the facilities available to teach this.

As a support centre, we offer drop in and on the spot support to students with specific queries, as an addition to the material learnt in lectures. However, our university is split across 3 sites, and as such it is not always possible for students to attend the additional sessions. There are also students who have not used statistical packages as part of their degree, but are expected to include a statistical component in their dissertation and therefore, we teach them from the basics upwards, targeting it specifically to their research and aims.

During the summer we have mainly Postgraduate level students and can deal with approximately 15-20 different research projects a week, in any field taught within the university. During semester time, we can deal with that in a day. So it can get a little busy. There are lots of aspects to the work. With the centre, it is separate from any of the course, but a way for students to access resources, and as such, we want to make sure that the students are being made aware of many different resources types.

The range of statistical tests on offer are ever increasing, if a student wants to learn a new technique that we have previously covered, then I work with that student to learn the technique and gain the understanding alongside them. This information is then added to the knowledge base for any future students to access it. Its a lot of work at times, but worthwhile when the results start coming in.

Statistics, in my opinion, is a skill-set to be learned as the need arises.  It may first arise when reading professional papers.  It may next arise when you do your own study, as a means of comparison.  So much can be done on line, or with the help of computerized statistical packages.  Yet, it still comes down to:  what do I need to know, and how can statistics aide me in my work.

I'm with Mr. Knox on this one, I've learned over the years that stats is best learned in real time as needed, usually accompanied by whatever stats package that will best aid with the work at hand.

https://www.khanacademy.org/math/statistics-probability

Thank you for the link M. Rose Barlow, there are some great videos and tutorials on that site.

In my doctoral studies, I found that statistics was nearly universally feared, but that the reality was far less intimidating. Mine was an online course wherein mastery had to be demonstrated, so it was challenging. I found that working with a group was helpful, but only when the group committed to learning statistical processes thoroughly and fundamentally. So, where possible, we broke models into individual processes and worked to understand those individual processes first, without necessarily relying on stats packages. It was illuminating, and when we took our problem solving to stats packages, both the processes and the results were much more easily grasped. Our group's shared anxieties helped to reduce our individual apprehensions.

Carol I teach foundational studies course on Elementary statistics. I hear your thiughts

 I start my course and teaching agenda introducing students with the concepts of quantitative literacy and statistics. The mathematical formula or notations equations they are learning and using is basically applied for real works situation. I use the concept of backward design to teach my statistics class. I always think what I want my students wto learn from this class. Based on that I reverse back and start working on the topics, design assessment tools. 

Majority of the time statistics class is taught with heavy dose mathematics without even realizing that it is necessary for the students to understand the science of statistics and how is it used in real life. For example y = mx + c is the basic algebra for regression equation where y is used as the dependent variable and x is the independent variable or explanatory variable. We can replace y with GDP ( gross domestic product ) and x can be replaced with savings. 

Using videos and hands on data set investigative task helps the students to understand the material.

khan academy videos are very helpful resources for students who are struggling to understand the mathematical concepts. 

In my doctorate courses, the research courses involving statistics did not benefit me probably because I have scant attitude for any math. I agree that instructors that make assumptions about the level of comprehension of statistics knowledge are a problem. Trying to focus lectures and coursework on real world applications might help learners like myself. I still dread statistics.

Wondering what anyone thinks about teaching statistical software programs like SPSS than statistics would be a better means to prepare learners for real world applications?

Learners were offered workshops outside of our programs. While nice, it was an online course without a facilitator. Also, time did not allow taking advantage of these workshops. Once we got to the analysis portions of dissertations, having skills with the software, in my opinion, would have better-served students.

Thank you all, as a support tutor, I don't teach the statistics modules built into the courses. What I do is help apply the analysis to the research. Although the students are taught the statistical packages in some courses, because the focus on running the tests in the software (spss for example), their questions to me tend to include, why do we do this and what does it mean though? The issue with focussing the teaching on statistical packages is that the students do not understand the reason for the test, they are just following instructions. The aim is to provide a range of resources that will help answer these questions and support the analysis of the data in programmes such as SPSS. I have also thought about offering webinars that will allow the user to join the online session from anywhere and still be able to speak to the facilitator if they have a question. We have three large campuses, so timetabled sessions or drop ins are not always accessible to all, so it is also about making the resources more accessible. We have gone from one centre to four in the last 2 years and are due to increase again soon, with full time tutors being available. The tutors are currently staff on zero hours contracts, so the centres are not open continually. It's starting to move in the right direction, do we want to keep the momentum going. 

As per your requirement, Webinar is the best option while interacting with the students at all the three campuses. You may consider dealing with the whys and hows of the statistical tests and more importantly the misconceptions regarding the selection of appropriate method and application of statistical tools. 

The main knowledge gap in students is generally seen in understanding the concepts of statistical techniques. The availability of statistical packages does not ensure its proper usage until one knows why a particular method is used and what it interprets. So I think the importance should be given more to conceptual clarity which helps in solving research problem. 

Based on my experience, students should be given early exposure on spss ... at least basic introduction to spss. There should be a basic test of calculations on the students involved. Demonstration instruction is necessary so students can see for themselves how data is included. For example: there should be a set of questions from the actual respondents. Lecturers who teach spss need to demonstrate the correct method for entering data .... usually spss lecturers continue to teach spss with circulation notes ... for more experienced will show how to label respondents' questions to make it easy for students to enter data.

Very valuable suggestions have been shared here. I also see Moodle as good platform for enabling the e-learning you had in mind.

But I would like to pause for a moment and ask what kind of learning we really want to enable. I am well-aware that my answer is not going to tackle the question entirely but hopefully it will provide additional value. 

In my opinion, applied statistics can be one of the most powerful tools to teach students critical thinking skills, which was ranked in the second place among the top-10 skills needed for us to thrive in the era of the Fourth Industrial Revolution, according to the World Economic Forum. Statistics can teach students, for instance, that correlation does not imply causation, the concept of statistical significance of results, and help them aptly design a research study.

In your case, the students might solve their exercise at home or in a lab and share their findings with the other course mates from a technical but also critical point of view, in the classroom or in a web-forum. 

Good luck!

Hi

It is one of the most challenge issues and need hardworking and must be patient.

Good luck.

I was involved in a series of projects funded by the Higher Education Academy which looked at the challenges of teaching of social sciences research methods— the projects had a quantitative methods and statistics emphasis. There are some ideas in this special issue dedicated to the projects:

Article Special Issue on Social Science Research Methods Education

Hi. I am currently teaching statistics at higher education to. After years of teaching, I never satisfied the way I promote thinking skills through introductory statistics content. As the environment here, lecture-centered is dominated, therefore it is challenging and require handwork to provide meaningful learning experiences rather than rote learning. Few ideas to share, I use info-graphic, newspaper article to tap into Why Statistics. We had a classroom discussion based on student's prior knowledge about statistics using materials provided. At least, before we introduce them with statistical tools, we help them to make sense of when and why do they need statistics in their real life applications. I would say, teaching statistics require "begin with end in mind" and constructivisim approach.