Identifying Suitable Data Sources

For any given research it is important to have the necessary data requirements so as to make sure that the experiment is undertaken successful. Consequently, prior to commencing on the experiment, the researcher is supposed to determine and identify suitable sources of data that the research will gather the needed information (Barker, and Milivojevich, 2016). Accordingly, data collection is the initial step when it comes to designing as well as implementing any given experiment. Certainly, after identifying the appropriate sources of data the researcher is supposed to gather raw data and then recorded it in well-organised tables for purposes of analysis at the time of implementing the experiment (Curtis et al, 2015, p. 3462). In this sense there are various steps that are undertaken in at the time of data gathering such as identifying the most suitable data sources, data collection and storage of the collected data.         

Accordingly, the initial step in any given research experiment, the primary most step is identification of appropriate sources of the data under investigation (Eyring et al, 2016, p. 1939). Therefore, it is highly advisable that the researcher determine the most appropriate type of data which is necessary for the experiment to avoid cases of collecting data which is not relevant with the research under investigation (Montgomery, 2017). Therefore, the type of data that will be collected will be used to show how business organisation utilise shopping apps to market and sale their products. The reason of picking of business organisation to be able to identify how these organisations work to ensure that customer’s personal information is protected.  Thus, it is significant for the researcher to select on the most suitable and dependable sources of data where it will be possible to interact with diverse personalities using different social applications to improve their personal data privacy (Candioti, De Zan, Cámara, and Goicoechea, 2014. p. 124). Certainly, some of the probable public place sources that are likely to provide the investigator with the appropriate data include malls, companies and small and medium-sized enterprises and supermarkets which run their business using social applications to interact with their existing and potential customers.          

Table 1: Data collection table  

After the data has been gathered and recorded in the acceptable way, the researchers is supposed to keep the gathered raw data in the most appropriate manner because this data will be needed in future. Therefore, some of the appropriate ways to keep this data is through the use of table which have to be saved and kept safe to ensure that these data is not interfered with or accessed by unauthorised persons who might corrupt it (Campbell, and Stanley, 2015). The raw data can be stored in the table as illustrated below.

Table 2: Data storage table  

When the data collection has been accomplished and the data safely stored, the following crucial step in the experiment is the design and implementation phase. During the design and implementation where the gathered data is modified as needed to ensure that it is possible to effectively utilise the data in the experiment (Chandrasekaran et al., 2017). As a result, this part involve a number of practices such as data pre-processing feature selection or dimension reduction, experimental design, and lastly but not least real time implementation of the experiment.       

Gathering Raw Data

During the data processing time the collected data is transformed and refined and then converted into more simplified form which are easy to understand and analyse. It can be said that the data pre-processing is conducted to aid the researcher in preparing the data to be utilised in the experiment so that it can realise all the necessary features which have to be processed while analysing the experiment. Accordingly, data pre-processing is an essential procedure in an experiment since it plays a significant part in enhancing the usability and readability of the collected raw data. Consequently, once the experimental raw data has been collected, it is advisable to perform data pre-processing prior to making use of the experiment. In that case, there are several approaches utilised to perform data pre-processing, but the most appropriate approaches include data integration, data cleaning, data reduction, and data transformation.

Data integration: This is a pre-processing approach whereby the collected raw data is improved and then changed into a suitable form or formats accordingly. Therefore, the main practices during this phase include smoothing of data, data normalisation, summing up of the data, as well as data generalisation.

Data cleaning: This is a pre-processing technique that involves normalisation of the data to substitute for some of the erroneous data values, reduction of the data noise and eliminating of the unnecessary inconsistency made during data coaction time. Certainly, data cleaning helps to ensure that the researcher come up with data which is less noisy, complete, and consistent. As a result, this makes it easier for the researcher to analyse the experimental results.     

Data reduction: Accordingly, this process entails simplifying the data by eliminating unnecessary and unwanted data while ensuring that the data is not altered so as not to compromise the quality and integrity of the data. Data reduction involves a range of techniques such as data discretisation, dimension reduction, data compression, and numerosity reduction. Indeed, all this approaches are significant in the sense that they aid to eliminate unwanted and unnecessary data contained in the raw collected data. As a result, it help to ensuring that the researcher only remain with accurate data which can be effectively used in the experimental process with much easy.            

Data transformation: It is pre-processing technique that combine data that seem to be related even though it can be originating from different sources so as to come up with a single and much dependable data that can be utilised as one in the experiment. Thus it makes the experiment process easier.

Accordingly, the figure below provides an illustration of a summary of the major data pre-processing techniques used in experiments.  

Feature selection is a technique used to eliminate certain features contained in the collected raw data to ensure that the researcher get the most appropriate features and highly dependable features (Zhou, Lu, and Fujita, 2015, p. 53). Thus, this make it easy for the researcher to use these features in the experiment. Accordingly feature selection aid in averting encountering with unnecessary data when analysing the data to ensure that the experiment is highly dependable. At times feature selection includes reducing the dimensions of the raw data which is the reason as to why it is at times referred to as dimension reduction (Prusa, Khoshgoftaar, and Dittman, 2015, p. 300). After the feature selection process of data pre-processing and dimension reduction have been carried out the data can be recorded as illustrated in the table below.

Recording Data in Organised Tables

Table 3: Feature selection/dimension reduction table    

During the experiment design phase it when the methodology is employed is selected and then applied effectively to the experiment. Thus, it is significant to select a methodology that will ensure that the experiment is performed in a smooth way with no difficulty.  

In the context of this experiment design the research chose on a hybrid methodology. The hybrid method is significant since it allow the investigator to gather and analyse both arithmetical and non-arithmetic data types. Therefore, it was one of the most effective approach for this experiment because the researcher anticipated coming across both arithmetical and non-arithmetic data (Haley et al., 2018, p. 4). The researcher was mainly interested in how shopping apps aided in enhancing the sales of business and how it effectively met customers’ needs and wants. As a result, the researcher paid visits to different public places where they could come into contact with this information. Consequently, the researcher developed some guiding research questions in form of a questionnaires. This questionnaire comprised of various levels such as the level of education, sex, age and background of the participants. Also, another section included how the participants utilised shopping apps in their smartphones. The researcher used both closed-ended questions and open-ended questions to allow participants to give their opinions regarding their experience with the shopping apps. For purposes of ethical considerations the research observed high ethical standards by ensuring that the questions asked could not infringed the respondent’s personal questions. Therefore, the research utilised statistical figures to record and analyse the collected data from the participants which are illustrated in the table below.

Table 4: The questionnaire questions table 

 What is your view regarding on the improvement that should done on the shopping apps to make them better? Give your recommendation.  

1. What are some of the challenges or problems that you face when using shopping apps to make your purchases?  What is your suggestions to better the shopping apps?

Certainly after developing the questionnaire, the investigator created a different table which was used to classify the participants based on their age, sex, education level and background. Indeed, classifying these participants was mandatory to allow the researcher to understand how shopping apps were utilised and how frequent this apps were used by diverse demographic data.

After developing the research questions and classifying the demographic population information of the respondents, the investigator went head to collect the data as asked by the experiment. The gathered data was then analysed with the help of certain statistical analysis techniques and tools with Microsoft Excel being the chief software for analysing data.

The table blow shows the analysed demographic population information for individuals who were involved in the research.

Table 6: Analysed demographic population information 

 The total number of respondents used in the research was 250   

The target research population used in this case involved 250 participants. The results that after circulating the questionnaire to be filled in by respondents’ involved 250 respondents is lustrated in the table below.

Prior to conducting any research the investigator is supposed to have an anticipation regarding the likely research outcome. Accordingly, prior information regarding the research experiment plays a significant role in aiding the investigator to work within the experiment scope and avoid straying of the experiment. It also, make it easier to find know what is expected of the experiment. Therefore, to get the expected idea regarding the study is found from past research on a similar research subject literature which the researcher intent to expound their understanding. Thus, before performing the research the researcher is required to ensure that there is a sufficient number of people using smartphones and they are also aware of the existence of the shopping apps in their cell phones. Also, some of the individuals should be already using shopping apps to make purchases to enhance their purchasing experience.         

The anticipated results was not far much from the real time experimental outcomes that were attained after performing the research. After undertaking the research the researcher found the following results:

Out of the 250 participants 245 participants were in possession of smartphones with shopping apps.

92% of the respondents were conversant with the shopping apps on their smartphones.

Among the research sample 75.6% had already downloaded the shopping apps and using them.

60% of the respondents who took part in the research found that shopping apps were effective and enhancing their purchasing experience.

68% of the respondents were satisfied with the performance of the shopping apps in making purchases.

Out of the 250 participants only 32% gave a recommendation regarding improvement on the shopping mall.

Lastly, only 6% of the respondents experienced challenges with using the shopping apps, which means that the shopping apps were highly effective in making purchases. 

Barker, T.B. and Milivojevich, A., 2016. Quality by experimental design. CRC Press.

Barker, T.B. and Milivojevich, A., 2016. Quality by experimental design. CRC Press.

Campbell, D.T. and Stanley, J.C., 2015. Experimental and quasi-experimental designs for research. Ravenio Books.

Candioti, L.V., De Zan, M.M., Cámara, M.S. and Goicoechea, H.C., 2014. Experimental design and multiple response optimization. Using the desirability function in analytical methods development. Talanta, 124, pp.123-138.

Chandrasekaran, A., Anand, G., Ward, P., Sharma, L. and Moffatt-Bruce, S., 2017. Design and Implementation of Standard Work on Care Delivery Performance: A Quasi-Experimental Investigation.

Curtis, M.J., Bond, R.A., Spina, D., Ahluwalia, A., Alexander, S.P., Giembycz, M.A., Gilchrist, A., Hoyer, D., Insel, P.A., Izzo, A.A. and Lawrence, A.J., 2015. Experimental design and analysis and their reporting: new guidance for publication in BJP. British journal of pharmacology, 172(14), pp.3461-3471.

Eyring, V., Bony, S., Meehl, G.A., Senior, C.A., Stevens, B., Stouffer, R.J. and Taylor, K.E., 2016. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geoscientific Model Development, 9(5), pp.1937-1958.

Haley, N.J., Richt, J.A., Davenport, K.A., Henderson, D.M., Hoover, E.A., Manca, M., Caughey, B., Marthaler, D., Bartz, J. and Gilch, S., 2018. Design, implementation, and interpretation of amplification studies for prion detection. Prion, pp.1-10.

Merriam, S.B. and Tisdell, E.J., 2015. Qualitative research: A guide to design and implementation. John Wiley & Sons.

Montgomery, D.C., 2017. Design and analysis of experiments. John wiley & sons.

Prusa, J.D., Khoshgoftaar, T.M. and Dittman, D.J., 2015, May. Impact of Feature Selection Techniques for Tweet Sentiment Classification. In FLAIRS Conference (pp. 299-304).

Zhou, L., Lu, D. and Fujita, H., 2015. The performance of corporate financial distress prediction models with features selection guided by domain knowledge and data mining approaches. Knowledge-Based Systems, 85, pp.52-61.