Mixed Methods Research Design For Determining Opinions On Online Shopping And E-business

Data Collection

The data to be chosen will be qualitative data to help determine various opinions on online shopping and e-business. The qualitative data will be collected either through focus groups, personal interviews as well as open-ended questions, (Gill et.al, 2008).

Another type of data to be chosen will be quantitative data where questions on age, time, and cost will be amongst those addressed. Quantitative data will be collected through Outcome Measurement Systems (OMS) questionnaires, (Dudoviskiy, 2016)

The approach research method will be mixed methods. Specifically, a sequential mixed methods design will be taken into consideration. This will allow for an initial qualitative stage of data collection and analysis followed by a stage of quantitative data collection and analysis and finally integrating both results, (Berman, 2017).

Types of data:

According to Sharma (2020), the types of data are inclusive of ordinal, nominal, discrete, and continuous data. Nominal data will be collected under qualitative data and will be inclusive of respondents’ gender. Continuous data will be collected under quantitative data and will be inclusive of respondents’ age.

The kinds of data to be collected will be written opinions to be answered in the questionnaires and open-ended questions.

Research Participants

The research participants from whom data will be collected will be inclusive of male and female gender in Vancouver. The participants are to be categorized on an age basis, where there will be 16 – 35 years and 36 -60 years. A total of 300 participants is to be taken into consideration. The data applied in the case of this research doesn’t change with time.

The biases to be taken into consideration will be: during the analysis of the data, one of the biases that are most likely to arise is the confirmation bias alluding to the fact that consumers definitely prefer online shopping. This bias will be overcome by correctly analyzing the questionnaire responses exactly as the given opinions. Another bias that may tend to arise is the assumption that young people tend to buy more online than elder people. This bias will be overcome by collecting 50% of the data from the older people and 50% of the data from the younger people.

According to Ailen (2022), the ethical issues that will be addressed are inclusive of the following: web tracking which refers to the act of collecting an individuals’ information since once a person visits a site their internet history is left there thus web trackers use applications to download and access this information to use it for their own benefit. Online privacy can be related to an example where monitoring systems are installed in an organization so as to track the emails sent and received by employees to enable determine if they are work associated or not. Web spoofing refers to where individuals create a fake website similar to the original one to lure customers to give their personal information. Cyber-squatting refers to buying and using another well-known organization’s domain name so as to infringe its trademark. Email spamming refers to sending unauthorized emails to people to lure them to enter their personal information and bank card numbers on the fake website. 

Types of Data

Analysis of Stationary data

According to Philips et.al (1992) stationary data doesn’t depend on the time at which the data was collected. This means that it is not affected by time changes.

The method that will be used to analyze stationary data will be use of differencing approach to calculate the mean and variance of each partition to see if it changes considerably over time.  According to Philips, the process of differencing is used to achieve a stationary mean and variance. The differencing approach has first order, second order and seasonal differencing. Philips applied the seasonal differencing which shows the difference between an observation and a previous observation in the same season, the Random Model is y`_t = y_t – y_t –_m, where m = number of seasons. Philips results were as follows: 

The above table according to Philips represents the logarithm table that helps to transform data using logarithms then seasonal differences are calculated. 

The table above according to Philips shows seasonal differences logarithms of the monthly scripts of A10 (antidiabetic) drug sold in Australia. 

The advantage of this differencing method according to Philips et.al (1992) is that it helps stabilize the mean of a time series by getting rid of changes in the levels of a time series thus reducing trend and seasonality. A shortcoming of the differencing method according to Laiser (2014) is that differencing cannot be used when the data appears to randomly fluctuate around its mean.

Validation of findings and conclusions will be done by ensuring the concluding statement and findings presented concur directly with the data results that have been analyzed. On the other hand the validity of the findings will be enhanced by ensuring the findings concur with the real situation in Online Shopping. Validity of findings according to Paparoditis (2010) can be enhanced by application of a varying covariance structure.

Analysis of Time Series Data

The analysis will be effectively done by carrying out multivariate time series data analysis on the time series data by creating multiple time series in a single chart, (Mary, 2021).  This method of analysis was previously used by Mary Terence. The example is as illustrated below: 

Mary applied the R – code to generate the above and below tables. 

The graph above represents data of total positive cases and total deaths from COVID – 19 weekly from 22 January 2020 to 15 April 2020 in data vector. 

Research Participants

The advantage of multivariate time series analysis is that the trend of multiple data can be shown at a single view and forecasting of data is possible, (Alam, 2020). The shortcoming of multivariate time series analysis is that it is a very complex task and thus requires a lot of effort, (Pena and Sanchez, 2007).

The validity of the conclusion and findings is done by carrying out a forecasting test using the Python – code to observe the actual and forecasted values of the ARIMA model, the values are then plotted in a line plot to show the results of the model, (Tiwari, 2020).

Hypothesis 1: A considerable number of individuals are perceived to buy products online since its more convenient.

Null Hypothesis Ho: Online shopping is convenient to customers.

Alternative Hypothesis H1: Online shopping is not convenient to customers.

 This hypothesis has been previously used by authors such as Ali and Karim (2010). The hypothesis will be tested using the Frequentist approach since through collecting data using questionnaires, certain opinions will recur in frequencies. The frequentist approach is best applicable in situations where frequencies arise, (Eser, 2021).

Timetable. The table below shows the different stages of the project that will be accomplished against a given period of weeks. The shaded region shows the number of weeks a stage will take to be completed.

Project Budget. The following table shows the budget for the 44 weeks. In general the expenses which shall be incurred during the project execution are shown in this budget.

Project Period

44 weeks/ 11 months (2022)

Budget for 11 months (2022)

References

Gill, P., Stewart, K., Treasure, E., & Chadwick, B. (2008). Methods of data collection in qualitative research: interviews and focus groups. British dental journal, 204(6), 291-295.    

Dudovskiy, J. (2016). The ultimate guide to writing a dissertation in business studies: A step-by-step assistance. Pittsburgh, USA, 51.

Berman, E. A. (2017). An exploratory sequential mixed-methods approach to understanding researchers’ data management practices at UVM: Integrated findings to develop research data services.

Up Grad. Sharma. R (2020, December 1)4 Types of Data. https://www.upgrad.com/blog/types-of-data/

UK Assignment Club. Alien. T (2022, March 18). Ethical issues in E-Commerce. https://www.assignmentclub.co.uk/blog/999-ethical-issues-in-e-commerce

Kwiatkowski, D., Phillips, P. C., Schmidt, P., & Shin, Y. (1992). Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root? Journal of econometrics, 54(1-3), 159-178.

Stack Exchange. Laiser. M (2014, January 14). Differenced data. https://stats.stackexchange.com/questions/82202/pros-and-cons-methods-for-detrending-time-series-data

Paparoditis, E. (2010). Validating stationarity assumptions in time series analysis by rolling local periodograms. Journal of the American Statistical Association, 105(490), 839-851.

GeeksforGeeks. Mary. A (2021, December 16). Time series analysis in R. https://www.geeksforgeeks.org/time-series-analysis-in-r/#:~:text=Time%20Series%20in%20R%20is,as%20given%20by%20the%20user.

Peña, D., & Sánchez, I. (2007). Measuring the advantages of multivariate vs. univariate forecasts. Journal of Time Series Analysis, 28(6), 886-909.

Towards Data Science. Alam. M (2020. April 10). Multivariate time series forecasting. https://towardsdatascience.com/multivariate-time-series-forecasting-653372b3db36

Towards Data Science. Twari. A (2020, August 30). Time-series Analysis, Modelling and Validation.

https://towardsdatascience.com/time-series-analysis-modeling-validation-386378cd3369

Eser, M. T. (2021). Comparison of Bayesian and Frequentist Factor Analysis Methods: Buss and Perry Aggression Questionnaire Example. International Online Journal of Education and Teaching, 8(4), 2871-2887.

Ali, M. M., & Al Karim, R. An Attempt to Identify the Key Factors Affecting Online Shopping Behaviour.

Order an Essay Now & Get These Features For Free:

Turnitin Report

Formatting

Title Page

Citation

Outline

Place an Order