A Handbook_NRA - Extended

Published in SCOPUS: Nik, Najeeb, & Asiah 51

Published in SCOPUS: Nik, Syed, & Norazlina 52

Under Review ISI: Nik, Rahimy, & Suraya 53

Under Review ISI: Haslinda & Nik 54

Under Review ISI: Najeeb & Nik 55

Recent Workshop: Suryo, Piveen, & Nik Does Merger & Acquisitions effect Attitudinal and Behavioral Loyalty? A Contingent Valuation Model on the Role of Involvement, Value, and Status Suryo & Piveen developed this Model in just 1 Day! 56

Published Book Chapter: Nik & Ali Al-bada 57

Published Book Chapter: Nik, Zafirah, & Manal 58

Letter of Acceptance: Premier Journal 59

You’re NEXT in line to SAY… I got Published! ... I got my PhD! -- First Step -- Thank you for buying this book… Handbook on the Novel Research Approach Your Path to Academic Success -- Second Step -- Join the research workshops by GradEx Academy… Preview & 3-day Novel Research Workshop -- Third Step -- Get advise from GradEx Academy on your thesis and journal papers… Thesis Consultation & Coaching *Let’s help others – share this information with your Friends, Masters, and PhD students. 60

Decision is in YOUR hands! So far, we have only touched on the basic concept of the NRA… only the surface of the ice-burg. There are a lot more technical details that need to be understood. The succeeding Chapters will clarify these… 61

Chapter 3 Assessing Models & Hypothesis Development Stage 2: Applied Novel Research Section 62

“To Understand and Succeed…observe the Patterns” Let’s focus on distinguishing the major patterns in scholarly research, particularly with regard to conceptual models and hypo structures.

Highlights of this Chapter  Critical Assessment of Models (Simple, average, sophisticated, those to avoid)  Patterns in Model & Hypo Structures (Unfolding the standard academic practice)  Concerns with Moderators (How to overcome them)  Finding your Research Topic (How to do this in 30 minutes)  Designing your Framework (Utilize your blueprints)  Variable Selection Criteria (To safeguard your research) 64

The 1. Going beyond Main Effects Importance of • we typically say, “it depends….” Moderators? (Recap) 2. More Complex Models • “If we want to know how well we are in Only when both aspects are social science, an index that will serve us understood — namely, what IVs well is how far we have advanced in our lead to the DV, and when the IV- understanding of moderator variables in DV relationship is enhanced or our field.” (Hall & Rosenthal) attenuated — a firm will be able to make an informed decision for 3. Managerial Implications • found sig relationship between IV & DV its strategic actions. (the WHAT?). Is this good enough for managers to make a decision? • need to know the conditions that will make the IV & DV relationship “stronger” or “weaker” (the WHEN?) 65

If you want to know how 1. Find 3 Conceptual Models with Moderating these tasks/questions can be Hypotheses (from 3 good journals in addressed effectively, please Marketing/Management) refer to GradEx Academy’s 2. Extract their Models and all their Hypothesis website and search for Sentences on 3 separate pages (one for each workshop Video Recordings: article) Number 3. 3. Examine all three models & hypos in terms of sentence structure & clarity. Are they Task 2 different from each other? (30 minutes) 4. Which model do you like most? Have a look at the next 5. Show the 3 models & hypos (just DESIGN... couple of slides… they will give you a general NOT content) picture of the type of ------------------------------------------------------------- models to avoid and those Note: Avoid models that are too complicated & to consider. models with many mediating effects! 66

Example: Model 1 Find 3 moderating models, randomly. Try to compare the models that you found. Note: If you are working on your own thesis or paper, find 8 conceptual models with moderating hypotheses. 67

Example: Hypotheses 1 H1: Visual packaging will be positively related to (a) consumer Find 3 hypo structures interest, (b) purchase decision, and (c) repurchase intention. like this. When you list down their hypos, you H2: Informative packaging will be positively related to (a) consumer will be able to see a interest, (b) purchase decision, and (c) repurchase intention. pattern… try to compare the hypos you found. H3: Functional packaging will be positively related to (a) consumer interest, (b) purchase decision, and (c) repurchase intention. Note: If you are working on your own H4: The positive relationship between all packaging attributes and thesis or paper, find at consumer buying behavior (interest, purchase, repurchase) will be least 8 articles with strengthened when product-personality congruence is high. moderating hypotheses. H5: The positive relationship between all packaging attributes and consumer buying behavior (interest, purchase, repurchase) will be strengthened when consumer innovativeness is high. 68

Complicated Model: Avoid! This model looks at multiple IV-DV-Mediating relationships. Though this is considered as one method for model development, it will be more challenging to justify write-ups with “elaborate” variable relationships. The NRA method, however, focuses its attention on moderating effects. So, just keep searching for moderating models. Moderator terms or keywords (e.g., interaction effects, contingency) provided in the Intro section may come in handy. 69

70

Complicated Model: Avoid! A moderator variable is considered here but the model is quite complicated because it has 3 different stages. 71

Complicated Model: Avoid! Another complicated model with moderating effects. This model contains too many variables and is not practical for empirical investigation. This one has 16 variables… but acceptable because it’s a conceptual piece. Note: Ideally, an empirical paper or thesis should have around 6 to 8 variables. 72

Complicated Model: Avoid! Don’t be deceived… This model may look simple but it combines both moderator and mediator relationships in a single model. Nowadays, with the application of PLS-SEM, more and more researchers are introducing moderated- mediated frameworks. 73

74

Moderating DESIGN: Simple Question for Thought: Will this model be good enough for a PhD thesis? 75

Moderating DESIGN: Simple Question for Thought: Can a moderator variable also be conceptualized as an independent variable? Hint… Quasi Moderator 76

Moderating DESIGN: Simple Question for Thought: Is it sufficient enough to have only one IV in a conceptual model? Hint… Interactions 77

Moderating DESIGN: Simple Questions for Thought: What about moderators, can we have only one MV in a conceptual model? Is this acceptable? 78

Moderating DESIGN: Simple (Nik & Lee) Question for Thought: Can authors decide NOT to show the model if it looks too simple like this one? 79

Moderating DESIGN: Average Questions for Thought: Guess how many hypos this model has? For models with multiple DVs, will data analysis be too heavy – i.e., drain statistical power? 80

Moderating DESIGN: Average Questions for Thought: 81 Do you know what multicollinearity is? Do you think this model has this problem?

Moderating DESIGN: Average Question for Thought: 82 Would demographics be a good choice for selecting moderating variables?

Moderating DESIGN: Average Questions for Thought: What are control variables? Are controls necessary? Should they be analyzed and hypothesized as well? 83

Moderating DESIGN: Average (Jahan & Nik) Question for Thought: Is this a mediating variable? This time I’ll give you the answer… This study seems to conceptualize mediating effects, but note that it’s not compulsory to test such effects; it all depends on your research objectives and how you argue your case. Most of the moderating designs shown so far are sufficient for a PhD thesis or journal paper, but an average model design will have a greater chance of getting accepted into a reputable journal. 84

Let's Discuss Task 2 1. Show the 3 Models with Moderating Hypos (DESIGN only, not content) 2. Are the hypos different from each other? 3. Which model do you like most? To thoroughly understand this Chapter and get an idea of how to address these tasks/questions (including Questions for Thought), please refer to GradEx Academy’s website and search for workshop Video Recordings: Number 3. 85

Example (for Task 2) 3 Conceptual Models with their respective Hypotheses… 86 Note: Based on the articles you found, which model & hypo structure do you like most?

Academics are always Difficult to get sig results. Even if there are sig concerned about developing effects, usually they are small (p < .10). So, how models with moderating can we overcome this problem? effects, especially one with multiple moderators. The 1) Select variables that are logically related (context) actual concern comes much later in the research 2) Acquire adequate sample size (minimum 120) process… the data analysis. Concerns with 3) Use ONLY existing measurements (Alpha > .70) Moderators 4) Limit the number of interactions (statistical power) • Many IVs + Few MVs (good) Here are some remedies to lessen • Few IVs + Many MVs (good) the burden of doubt. The details will be • Many IVs + Many MVs (avoid) explained latter in the NRA process. Based on my experience, out of I would not recommend models with Many IVs + Many approximately 80 data analyses that I have performed (with students), only MVs because you will end up with too many interaction one data did not produce significant moderating effects; and this was due to effects & hypos, which will then reduce statistical power. questionnaire structure, not because of But some students are ‘ambitious’ so we would normally the inclusion of moderating variables. resort to Hierarchical Regression Analysis… 87

Continue with We have covered simple and average designs, Moderating now let's take a look at…. Designs… • Higher-end Models and • Sophisticated Hypo structures … for your Exposure … pay attention to the PATTERNS Note: The models shown in the succeeding pages are more technical and requires care in their construction. These conceptual models can get easily confused or misunderstood and are harder to justify (e.g., 3-way Interactions). Unless you are an experienced researcher, I will not recommend pursuing these designs. 88

Moderating DESIGN: Sophisticated Question for Thought: Can you detect the focal construct in this model? 89

Academic Talent 90

Moderating DESIGN: Sophisticated Questions for Thought: Is this a Mediation model? This model clearly shows that MVs may also act as IVs, but can a MV act as a Mediator variable at the same time? 91

Moderating DESIGN: Sophisticated Questions for Thought: Can you identify their IVs? This model has one MV… or perhaps more? 92

Moderating DESIGN: Sophisticated (Enam & Nik) Question for Thought: Are the MVs separated or grouped together? 93

Moderating DESIGN: Unique Questions for Thought: How many IVs does this model posses? What about MVs? 94

Moderating DESIGN: Unique Question for Thought: How is this model uniquely different to the other ones I have shown? 95

Moderating DESIGN: Unique (Najeeb & Nik) Question for Thought: Can you spot the Moderator Variable(s) in this model? 96

Moderating DESIGN: Ultimate Questions for Thought: Do you think this is a moderated-mediated model? Why or why not? 97

Moderating DESIGN: Ultimate (Nik & Bryan) Question for Thought: How can we differentiate MVs & IVs? 98

Direct 1. (IV) is positively related to (DV). Hypos: 7 Different 2. (IV) is positively associated with (DV). Structures ---------------------------------------------------------- 3. (IV) will have a positive effect on (DV). 4. (IV) has a direct positive impact on (DV). ---------------------------------------------------------- 5. (IV) enhances the effect of (DV). 6. Increase in (IV) increases (DV). 7. The higher the (IV), the higher the (DV). Self-assessment: Based on the list of direct hypos you gathered, which pattern appeared the most? 99

Examples: Direct Hypos in Action 1. Innovativeness of a new product (IV) is negatively related to product quality (DV). 2. Perceived usefulness (IV) has a direct positive impact on the attitude of customers (DV). 3. Innovative culture (IV) enhances the effect of leadership (DV). 4. Increase in attitude towards private brands (IV) decreases customer intention to buy manufacture brands (DV). 100