Business Research Methods

Project
On
Conjoint Analysis
Identifying Key Product Attributes & Product Designing of Mobile Phones

Abstract

This paper intends to explore consumer preferences for Mobile phones attributes, to determine the optimal combination for consumers, and to provide manufacturers a reference for their marketing strategies. In this study, consumers were divided into several demographics (age, gender, occupation) and individual preferences for various mobile phone attributes were compared. Consumers showed significant demographical difference in their preferences over the combination of mobile phones attributes.
The various combination of mobile phone attributes were grouped together. Subjects were asked to rank the 22 product profiles (Pair1 to Pair22) from the most to the least preferred. The variables Pref1 through Pref7 contain the IDs of the associated product profiles, that is, the card IDs. Subject 1, for example, liked pair13 most of all, so PREF1 has the value 13.
Analysis of the data is a task that requires the use of command syntax—specifically, the CONJOINT command. The necessary command syntax has been provided in the file conjoint.sps.

Literature Review

N. Soutar et al. (2008), the study aims to examine its relevance in exploring the trade-offs followers make about leaders. The aim was to have an integrated understanding of leadership. The leaders were assessed on the eight leader attributes obtained from three focus groups. The study support the usefulness of conjoint analysis in leadership research as the results provided a clear picture of the way followers perceived their leaders. The findings of the study was that the participants traded off leader attributes sensibly, providing useful information about the attributes’ value. The value of the leadership attributes examined and the most valued leader attributes found were inspiration, trust and communication.

Marjon van der Pol et al. (1966), the purpose of this research paper is to know the factors affecting the preference of the consumers for the consumption of fruits and vegetables. The methodology used in this paper is the pilot questionnaire to know the factors affecting preferences for consumption of food and vegetables of consumers and most commonly fruits and vegetables consumed through conjoint analysis. The sample size was 600 that are randomly chosen households and the response rate was 63.5%. The result shows that there are 8 factors that affect the consumer preference of buying fruits and vegetables and the most important was price followed by taste and family preference.

Byoungho Jin et al. (2009), the purpose of the paper seeks to fill a research paper that aims to compare evaluative attributes that Chinese and Indian consumers utilize when purchasing denim jeans. The research methodology is such that the data were collected in Shanghai, China and Bangalore, India. The findings says that the attributes ranked by Indian and Chinese market were different. Chinese consumers placed the highest importance on price, followed by fitting, brand country of origin, quality, and design whereas Indian consumers placed importance on fitting, brand Country of origin, design, price, and quality, in descending order.

Anna Codini et al. (2012), the purpose of the paper seeks to fill a research gap that concerns empirical studies on value-based pricing in durable consumer goods. It analyses the relationship between value for the customer and market prices in the washing machines market. The research methodology is such that the customer value of a sample of 129 washing machine models is assessed through the conjoint analysis technique. It is then compared through a regression analysis to the market prices of the products. The conclusion follows as the regression analysis reveals that the alignment between price and value for the customer is limited (only one of the two subsamples presents a positive dependence among the variables).

Panoraia Papanagiotou et al. (2012), the analysis of the paper is to show the study on t investigating consumers’ pork meat quality perception, focusing on the stage prior to consumption. Furthermore, market segmentation is attempted on the basis of specific socio-demographic characteristics, pork consumption habits and quality perception. The research methodology used is such that a field research was performed in a Greek urban area, Thessaloniki. The research was based on consumers’ quality evaluations of pork meat pictures, which were computer-manipulated to acquire different levels of two intrinsic quality cues (colour and Marbling), while labels were added indicating various levels of two extrinsic quality cues (price and origin). Data were analyzed by means of conjoint analysis. The conclusion is that Greek consumers appear to attach more importance to marbling when forming pork quality judgments. Price is classified second in importance closely followed by colour, while country of origin appears to be the least important factor affecting quality perception of pork.

Stefan Mann et al. (2012), the paper investigates on determining the importance of the “organic” attribute for wine in relation to other attributes and to identify the most important determinants as far as the consumption of organic wine is concerned. The research methodology used as a survey based on interviews with 404 Swiss wine drinkers was conducted. The data were evaluated by means of conjoint and regression analyses. The conclusion is is such that the “organic” attribute was more important than the colour of the wine but less important than the price and the country of origin, foreign French wine being preferred to local Swiss wine. Urban and female consumers were relatively likely to consume organic wine, as were consumers who perceive organic wine as being healthy.

Barone and Lombardo (2004), the analysis of the paper seeks to fill a research paper showing judgement about different scenarios (specific combination of elements) for evaluation of relative importance about service elements composing a service using conjoint analysis. Design of experiment methodology is usually exploited.. In the findings, Abandon rate proved extremely to be reduced.

Marija Cerjak et al.(2010), the research paper investigates to determine, via an empirical study of beer consumers in Croatia, the influence of tasting on the validity of conjoint analysis (CA) under presence of familiar or Unfamiliar brands. The research methodology comprised a face-to-face survey with 403 beer consumers dividing the respondents into four groups using conjoint analysis. .Additionally, a structured questionnaire was used consisting of a few questions regarding respondents’ socio-economic characteristics, beer purchasing, and consuming behaviour. The findings of the study indicate that tasting should be used as a stimulus presentation method for CA with food and beverage products/brands, which are unfamiliar to the consumers

ULENGIN (1998 ), the purpose of the paper is to know the increased interest in the application of marketing concepts to retail banking, topics that generated much interest include the measurement of perceived service quality & modelling of consumer preference for banking products. Conjoint analysis is used to measuring consumer’s trade offs among product attributes, including service attributes & service dimensions. The present article has twofold. First we wish to demonstrate the implementation information approach that uses integrated experiment .secondly it aim to use service dimensions as defined in the original SERVQAL instrument

OSTROM and ICCOBUCCI (1995), in this research paper researcher used conjoint analysis & integrated experiment . first they make a hierarchy decision structure assumed in the method of integrated conjoint experiments. The provide four dimensions such as accessibility to the bank, competence to personnel , accuracy and friendliness & tangibles . Than they define some attributes of it. They used to select 27 profiles for each of the experiments .than they use regression technique. The approach assumes a hierarchy that maps attributes on to service dimensions. Regression model indicates that there are several dimensions that had different effects across experiments .

DE SARBO et al. (1993), in this paper they use of integrated conjoint analysis as a method for studying the relations between attributes that determines perceived service quality ,dimensions of perceived service quality dimensions and overall performance for the service.The approach assumes a hierarchy that maps attributes on to service dimensions./ which in turn determine the overall preference of the service. A major advantage of this approach is that the number of attributes can be much larger than is possible in conventional conjoint approaches

GREEN and SPRINIVASAN (1978), the purpose of the paper is to analysis female consumer’s apparel purchasing decision based on the value they attach to certain attributes when making a purchasing decision the research methodology in past research mostly used survey methods that directly measure consumer’s attitude towards product & services. Churchill and Iacobuccialso describe a causal research design as one that is concerned with cause-and-effect relationships. Studies of this kind normally take the form of experiments, because they are best suited to determining cause and effect. Sudman and Blair group experiments as one of the three major sources of primary data (the other two being surveys, and focus groups and depth interviews).

Cooper and Schindler (1998), past research mostly used survey methods that directly but in this research paper they measure consumers’ attitudes towards products and their attributes. use of experiments as a source of primary data. In this research paper researcher have different viewpoints with respect to classification value and use of experiments such as a sources of experiments as a sources of primary data . they use eight different descriptors to classify a research design. One of these is the power of the researcher to manipulate the variables, in which case two kinds of design can be used, namely experimental and post facto.

Churchill and Iacobucci ( 2002), conjoint analysis places more emphasis on the ability of the researcher or manager to theories about the behaviour of choice than it does on analytical techniques. Thus it should be viewed primarily as exploratory, because many of its results are directly attributable to basic assumptions made during the course of the design and the execution of the study. The critical interplay between the assumed conceptual model of decision-making and the appropriate elements of the conjoint analysis makes this a unique multivariate method .researcher taken following steps first they select attributes than determine attribute level than attribute combination , select form of presentation of stimuli and nature of judgement , decide on aggregation of judgments than finally select the analysis technique

Darian and Tucci (2011), the purpose of this research is to interrogate the health benefits of foods to the consumers and significance of these benefits. The approach or methodology of this research is that a statistical technique called conjoint analysis is used to interrogate the significance of five major health benefits to consumer which affects their buying decisions. The sample sizes that are people who are interrogated are 238.The conclusion drawn from the research is that the high nutritional value is the most significant health benefit that affects the buying decision of the consumers and if two benefits are to be promoted than the best combination will be high nutritional value and the potential to reduce cancer and next would be high nutritional value with proven to reduce the risk of heart disease.
Warneke and Schneider (2011), the research investigates to outline an approach which is verifiable to convey the preference of the employees to their employer thinking for foreign assignments in abroad. The methodology is to use conjoint analysis and Utility theory to check the usefulness linked to the assignments which are working condition, career prospect and living prospect etc are the employees preferences. Preferences like specific status quo may be measured through conjoint analysis. The conclusion from the research states that the policies and procedures for compensation package of expatriate may be formulated to reduce the gap between the standardization and the preferences of expatriates working worldwide.
Ernest North et al. (2002), the study of the paper is to know the application of conjoint analysis which is a tool to understand how respondents develop preferences for products or services. The research methodology which is used to understand the conjoint analysis is the laboratory experiment. In laboratory experiment respondents are asked to make judgments about their preferences for various attribute combinations during purchasing of apparel. The conclusion of the experiment is that it helps to understand the behaviour of of consumers when making purchasing decisions in retail situation.
Paul E.green et al. (1990), the purpose of the paper is to understand the technique for measuring buyer’s trade off among multi attributed products and services and also to know about the useful avenues for new research. The research methodology is that 400 commercial applicants per year interviewed to collect the information. Then least square regression method is used to find the relation among the large number of attributes and levels within the attributes. The conclusion of the study reveals that various levels of attributes leads the biases in attribute importance.
Feurtein et al. (1999), the purpose of the paper is to understand the choice based conjoint analysis (CBC) software which is used in web based data based collection. The research methodology is such that here respondents’ expresses preferences by choosing concepts from set of concepts rather by rating or ranking. The conclusion is that it attracts marketing research field more because the task of choosing a preferred concept is similar to what buyers actually do in market place.

Methods

We have used Primary Source of data collection in our research. Primary sources are the sources from which information collected is first hand. In primary data collection, person collects the data himself using methods such as interviews and questionnaires. The key point here is that the data collected is unique and, until that person publish; no one else has access to it.
There are many methods of collecting primary data and the main methods include Questionnaires, Interviews, Focus Group Interviews, Observation, Case-Studies, Diaries, etc.

The primary method we have used in our research paper is Questionnaire. We have taken the responses of 150 respondents of different age group belonging to different backgrounds.

The design used in the research paper is Orthogonal Design.

Results ANALYSIS AND INTERPRETATION OF THE TEST:

ONE SAMPLET-TEST :
THE ONE SAMPLE t TEST : TESTING HYPOTHESES FOR SINGLE SAMPLES.

t = sample mean – population mean estimated standard error of mean (derived from sample) (notice similarity to Z. t = Z when sample size is large) * tests the null hypothesis that the mean of a particular sample differs from the mean of the population only by chance * Confidence intervals for a nominated level of probability provide a range of values within which the population mean is likely to lie. * Using the same sort of logic, a different approach is a one sample test. * In a one sample test, we have nominated a known or possible population mean and we have conducted the test to determine whether it is likely (with a nominated level of probability) that the mean value we have obtained from our sample could have come from the population with the nominated population mean(4.9155)

Interpretation:

Null Hypothesis : There is an independent association between the choices of people and any association is only a chance occurrence.

Alternate Hypothesis : There is significant association between the choices of people and there is strong likelihood of people towards a particular Brand.

Group Statistics | | GENDER | N | Mean | Std. Deviation | Std. Error Mean | SYMBIAN1 | 1 | 79 | 3.34 | 1.867 | .210 | | 2 | 71 | 3.11 | 1.617 | .192 | BLACKBERRY1 | 1 | 79 | 3.87 | 1.488 | .167 | | 2 | 71 | 3.38 | 1.367 | .162 | WINDOWS1 | 1 | 79 | 3.84 | 1.497 | .168 | | 2 | 71 | 3.90 | 1.385 | .164 | IOS1 | 1 | 79 | 4.19 | 1.798 | .202 | | 2 | 71 | 4.38 | 1.659 | .197 | IOS2 | 1 | 79 | 4.18 | 2.055 | .231 | | 2 | 71 | 4.06 | 1.874 | .222 | IOS3 | 1 | 79 | 4.37 | 1.806 | .203 | | 2 | 71 | 3.92 | 1.637 | .194 | ANDROID1 | 1 | 79 | 4.25 | 1.918 | .216 | | 2 | 71 | 4.08 | 1.888 | .224 | WINDOWS2 | 1 | 79 | 4.41 | 1.990 | .224 | | 2 | 71 | 4.18 | 1.877 | .223 | SYMBIAN2 | 1 | 79 | 4.51 | 1.860 | .209 | | 2 | 71 | 4.79 | 1.502 | .178 | BLACKBERRY2 | 1 | 79 | 4.58 | 1.722 | .194 | | 2 | 71 | 4.42 | 1.704 | .202 | BLACKBERRY3 | 1 | 79 | 5.00 | 1.577 | .177 | | 2 | 71 | 4.86 | 1.561 | .185 | ANDROID2 | 1 | 79 | 4.56 | 1.708 | .192 | | 2 | 71 | 4.56 | 1.910 | .227 | ANDROID3 | 1 | 79 | 3.99 | 1.918 | .216 | | 2 | 71 | 4.08 | 1.903 | .226 | ANDROID4 | 1 | 79 | 4.53 | 1.647 | .185 | | 2 | 71 | 4.38 | 1.477 | .175 | IOS4 | 1 | 79 | 4.09 | 1.923 | .216 | | 2 | 71 | 4.58 | 1.834 | .218 | WINDOWS3 | 1 | 79 | 4.53 | 1.535 | .173 | | 2 | 71 | 4.13 | 1.638 | .194 | BLACKBERRY4 | 1 | 79 | 4.29 | 1.868 | .210 | | 2 | 71 | 4.45 | 1.803 | .214 | IOS5 | 1 | 79 | 4.34 | 1.839 | .207 | | 2 | 71 | 4.56 | 1.962 | .233 | BLACKBERRY5 | 1 | 79 | 4.43 | 1.685 | .190 | | 2 | 71 | 4.46 | 1.472 | .175 | SYMBIAN3 | 1 | 79 | 4.54 | 1.584 | .178 | | 2 | 71 | 4.80 | 1.712 | .203 | WINDOWS4 | 1 | 79 | 4.57 | 1.795 | .202 | | 2 | 71 | 4.86 | 1.659 | .197 | IOS6 | 1 | 79 | 4.43 | 1.939 | .218 | | 2 | 71 | 4.37 | 1.701 | .202 |

Here, Symbian1 and Blackberry1 are the most preferred mobile phones among females and males respectively with P –value = 0.00 which is less than p <0.05 that further shows the significant association with respect to other 20 pairs. Hence, we reject the NULL HYPOTHESIS.

Independent GROPUP T-Test * The Independent Groups t test is used to determine whether two random samples are likely to come from the same population. * If there is a statistically significant difference between them then the null hypothesis that their sample means are simply two chance variations around a population mean is rejected. * Often used in the experimental versus control group set-up – the Between Subjects design.
Interpretation:
It tells whether the homogeneity of variance assumption has been satisfied or not. If p value is statistically significant then the variances are unequal and we must use the lower line of figures.
Here in case of Blackberry1 only the p value is significant, hence we have used the lower line of figures.

Group Statistics | | GENDER | N | Mean | Std. Deviation | Std. Error Mean | SYMBIAN1 | 1 | 79 | 3.34 | 1.867 | .210 | | 2 | 71 | 3.11 | 1.617 | .192 | BLACKBERRY1 | 1 | 79 | 3.87 | 1.488 | .167 | | 2 | 71 | 3.38 | 1.367 | .162 | WINDOWS1 | 1 | 79 | 3.84 | 1.497 | .168 | | 2 | 71 | 3.90 | 1.385 | .164 | IOS2 | 1 | 79 | 4.18 | 2.055 | .231 | | 2 | 71 | 4.06 | 1.874 | .222 | ANDROID3 | 1 | 79 | 3.99 | 1.918 | .216 | | 2 | 71 | 4.08 | 1.903 | .226 |

Here, it shows that in case of symbian1: Females prefer it more as compared to Males as their mean is more toward favorableness,Blackberry1: Females prefer it more as compared to Males as their mean is more toward favorableness,Windows1: Males prefer it more as compared to Females as their mean is more toward favorableness respectively.
Conclusion:

Conclusion

It can be concluded that the most preferred mobile phone is Symbian1 followed by Blackberry1.
The company must focus on the marketing strategies which highlights various key features like price and operating system to increase the sales of the lower mobile products and try to gain competitive advantage.

References

Jean C. Darian, Louis Tucci, (2011),"Perceived health benefits and food purchasing decisions", Journal of Consumer Marketing, Vol. 28 Iss: 6 pp. 421 - 428

Doris Warneke, Martin Schneider, (2011),"Expatriate compensation packages: what do employees prefer?", Cross Cultural Management: An International Journal, Vol. 18 Iss: 2 pp. 236 – 256

Stefano Barone, Alberto Lombardo, (2004),"Service Quality Design through a Smark Use of Conjoint Analysis", Asian Journal on Quality, Vol. 5 Iss: 1 pp. 34 – 51

Marjon van der Pol, Mandy Ryan, (1996),"Using conjoint analysis to establish consumer preferences for fruit and vegetables", British Food Journal, Vol. 98 Iss: 8 pp. 5 – 12

Feurstein, Markus and Martin Natter (1999), “Forecasting Scanner Data by Choice-Based Conjoint Models,” Sawtooth Software Conference Proceedings.

North and Vos, (2002), “The use of conjoint analysis to determine consumer buying preferences” ISSN 0378-5254 Journal of Family Ecology and Consumer Sciences, Vol 30

Harmen Oppewal, Marco Vriens, (2000),"Measuring perceived service quality using integrated conjoint experiments", International Journal of Bank Marketing, Vol. 18 Iss: 4 pp. 154 - 169

Geoffrey N. Soutar, Shaun Ridley, (2008),"Looking at leaders: a conjoint analysis", Leadership & Organization Development Journal, Vol. 29 Iss: 5 pp. 461 - 472

Stefan Mann, Ali Ferjani, Linda Reissig, (2012),"What matters to consumers of organic wine?", British Food Journal, Vol. 114 Iss: 2 pp. 272 - 284

Paul Hansen, Alison Hendry, Ray Naden, Franz Ombler, Ralph Stewart, (2012),"A new process for creating points systems for prioritising patients for elective health services", Clinical Governance: An International Journal, Vol. 17 Iss: 3 pp. 200 - 209

Anna Codini, Nicola Saccani, Alessandro Sicco, (2012),"The relationship between customer value and pricing strategies: an empirical test", Journal of Product & Brand Management, Vol. 21 Iss: 7 pp. 538 - 546

Green and Srinivasan(1990),"New development wuth implications for research and practice “, Journal of Marketing ,Vol. 13, Iss: 7 pp. 538 - 546

Marija Cerjak, Rainer Haas, Damir Kovacic, (2010),"Brand familiarity and tasting in conjoint analysis: An experimental study with Croatian beer consumers", British Food Journal, Vol. 112 Iss: 6 pp. 561 – 579

Byoungho Jin, Jin Yong Park, Jay Sang Ryu, (2010),"Comparison of Chinese and Indian consumers' evaluative criteria when selecting denim jeans: A conjoint analysis", Journal of Fashion Marketing and Management, Vol. 14 Iss: 1 pp. 180 -194

Geoffrey N. Soutar, Shaun Ridley, (2008),"Looking at leaders: a conjoint analysis", Leadership & Organization Development Journal, Vol. 29 Iss: 5 pp. 461 - 472

Appendices